Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L ...thedryneedlinginstitute.net/wp-content/uploads/2012/02/Dry... · Analysis 01.03. Comparison 01 acupuncture versusno treatment

Acupuncture and dry-needling for low back pain (Review)

Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L, Koes BW, Berman BM

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library2008, Issue 3

http://www.thecochranelibrary.com

1Acupuncture and dry-needling for low back pain (Review)

Copyright © 2008 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd

http://www.thecochranelibrary.com

T A B L E O F C O N T E N T S

1ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3CRITERIA FOR CONSIDERING STUDIES FOR THIS REVIEW . . . . . . . . . . . . . . . . . .

3SEARCH METHODS FOR IDENTIFICATION OF STUDIES . . . . . . . . . . . . . . . . . . .

4METHODS OF THE REVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5DESCRIPTION OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5METHODOLOGICAL QUALITY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10AUTHORS’ CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

10FEEDBACK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11POTENTIAL CONFLICT OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . .

11ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

11SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

12REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

16Characteristics of included studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

58Characteristics of excluded studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

59Characteristics of ongoing studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

62ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

62Table 01. Methodological quality assessment . . . . . . . . . . . . . . . . . . . . . . . . .

69Table 02. Clinical relevance assessment . . . . . . . . . . . . . . . . . . . . . . . . . . .

72Table 03. Adequacy of acupuncture . . . . . . . . . . . . . . . . . . . . . . . . . . . .

75Table 04. Improvement in pain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

76Table 05. Search Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

77Table 06. Internal Validity Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

78ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

78Comparison 01. acupuncture versus no treatment ((Sub)acute LBP: < 3 months) . . . . . . . . . . . . .

78Comparison 02. acupuncture versus placebo or sham intervention ((Sub)acute LBP: < 3 months) . . . . . . .

79Comparison 03. acupuncture versus other intervention ((Sub)acute LBP: < 3 months) . . . . . . . . . . .

79Comparison 04. acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months) . . . . . . . . . . . . .

79Comparison 05. acupuncture versus no treatment. (Chronic LBP: > 3 months) . . . . . . . . . . . . .

79Comparison 06. acupuncture versus placebo or sham intervention (Chronic LBP: > 3 months) . . . . . . . .

80Comparison 07. acupuncture versus other intervention. (Chronic LBP: > 3 months) . . . . . . . . . . .

81Comparison 08. acupuncture versus acupuncture. (Chronic LBP: > 3 months) . . . . . . . . . . . . . .

81Comparison 09. dry-needling versus other intervention ((Sub)acute LBP < 3 months) . . . . . . . . . . .

81Comparison 10. acupuncture versus placebo or sham intervention (unknown / mixed duration of low back pain) . .

81Comparison 11. acupuncture versus other intervention (unknown / mixed duration of low back pain) . . . . .

82Comparison 12. acupuncture versus acupuncture. (unknown / mixed duration of low back pain) . . . . . . .

82Comparison 13. acupuncture plus intervention versus other intervention alone. (Chronic LBP: > 3 months) . . .

82INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

82COVER SHEET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

84GRAPHS AND OTHER TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

84Figure 01. Acupuncture compared to no treatment, placebo or sham therapy . . . . . . . . . . . . . .

85Figure 02. Acupuncture compared to another intervention or added to other interventions . . . . . . . . .

86Figure 03. Effects of dry-needling at trigger points . . . . . . . . . . . . . . . . . . . . . . .

86Figure 04. Comparison between two techniques of acupuncture . . . . . . . . . . . . . . . . . . .

87Analysis 01.01. Comparison 01 acupuncture versus no treatment ((Sub)acute LBP: < 3 months), Outcome 01 pain .

iAcupuncture and dry-needling for low back pain (Review)


88Analysis 01.03. Comparison 01 acupuncture versus no treatment ((Sub)acute LBP: < 3 months), Outcome 03 functional

status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

89Analysis 01.04. Comparison 01 acupuncture versus no treatment ((Sub)acute LBP: < 3 months), Outcome 04 physical

examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

90Analysis 02.01. Comparison 02 acupuncture versus placebo or sham intervention ((Sub)acute LBP: < 3 months),

Outcome 01 pain (VAS) (lower values are better) . . . . . . . . . . . . . . . . . . . . .


Outcome 03 functional status (higher scores are better). Generic instrument . . . . . . . . . . . .


Outcome 04 physical examination: finger-floor distance (lower values are better) . . . . . . . . . . .


Outcome 06 mean difference in pain (final - initial) . . . . . . . . . . . . . . . . . . . . .


Outcome 07 mean difference in functional status (final - initial) Generic instrument . . . . . . . . .


Outcome 08 mean difference in physical examination (final - initial): finger-floor distance . . . . . . .

96Analysis 03.01. Comparison 03 acupuncture versus other intervention ((Sub)acute LBP: < 3 months), Outcome 01 pain

(VAS): lower values are better . . . . . . . . . . . . . . . . . . . . . . . . . . . .

97Analysis 03.02. Comparison 03 acupuncture versus other intervention ((Sub)acute LBP: < 3 months), Outcome 02

global measure (higher values are better) . . . . . . . . . . . . . . . . . . . . . . . . .


functional status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


physical examination (finger floor distance) . . . . . . . . . . . . . . . . . . . . . . .


Side effects / Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

101Analysis 04.01. Comparison 04 acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months), Outcome 01 pain .

102Analysis 04.02. Comparison 04 acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months), Outcome 02 global

measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

103Analysis 04.03. Comparison 04 acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months), Outcome 03

functional status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

104Analysis 04.04. Comparison 04 acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months), Outcome 04 physical

examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

105Analysis 05.01. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 01 pain

(instruments: VAS and number of words) . . . . . . . . . . . . . . . . . . . . . . . .

106Analysis 05.02. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 02 global

measure (improvement) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

107Analysis 05.03. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 03 functional

status (higher values are better) . . . . . . . . . . . . . . . . . . . . . . . . . . . .

108Analysis 05.04. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 04 limitation

of activity (higher values are worse) . . . . . . . . . . . . . . . . . . . . . . . . . .

109Analysis 05.05. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 05 physical

examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

110Analysis 05.07. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 07 functional

status (standardized measures) . . . . . . . . . . . . . . . . . . . . . . . . . . . .

111Analysis 06.01. Comparison 06 acupuncture versus placebo or sham intervention (Chronic LBP: > 3 months),

Outcome 01 pain (lower values mean better) . . . . . . . . . . . . . . . . . . . . . . .


Outcome 02 global improvement (higher values are better) . . . . . . . . . . . . . . . . . .


Outcome 03 pain disability index (lower values are better) . . . . . . . . . . . . . . . . . .


Outcome 04 physical examination (fingertips-to-floor distance).( Lower values are better) . . . . . . . .

iiAcupuncture and dry-needling for low back pain (Review)



Outcome 05 improvement in physical examination . . . . . . . . . . . . . . . . . . . . .


Outcome 06 Sick leave (higher values mean worse) . . . . . . . . . . . . . . . . . . . . .


Outcome 07 Well being (SF-36). (Higher values are better) . . . . . . . . . . . . . . . . . .


Outcome 08 Side effects / Complications . . . . . . . . . . . . . . . . . . . . . . . .


Outcome 09 pain (percent of baseline values) . . . . . . . . . . . . . . . . . . . . . . .


Outcome 10 sick leave . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


Outcome 11 general level of pain (0-15 points)(more points mean less pain) . . . . . . . . . . . .


Outcome 12 pain: difference between within group changes . . . . . . . . . . . . . . . . . .


Outcome 13 function: difference between within group changes . . . . . . . . . . . . . . . .


Outcome 14 Pain: percentage of patients with >50% pain reduction . . . . . . . . . . . . . . .


Outcome 15 spine range of motion: difference between within group changes . . . . . . . . . . . .

122Analysis 07.01. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome 01 pain

(lower values are better) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

123Analysis 07.03. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome 03 back

specific functional status (lower scores mean better). Ex: RDQ, Oswestry and Aberdeen . . . . . . . .

123Analysis 07.04. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome 04 back

specific functional status (higher scores are better). Ex: Japan Orthopedic Association Score. . . . . . . .


disability index (lower values are better) . . . . . . . . . . . . . . . . . . . . . . . . .

124Analysis 07.06. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome 06

physical examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


return to work (higher values mean better) . . . . . . . . . . . . . . . . . . . . . . . .

125Analysis 07.08. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome 08 Side

effects / Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


and function (adjusted for baseline values) . . . . . . . . . . . . . . . . . . . . . . . .


general level of pain (0-15 points)(more points mean less pain) . . . . . . . . . . . . . . . . .

127Analysis 07.11. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome 11 pain:

difference between within group changes . . . . . . . . . . . . . . . . . . . . . . . .

128Analysis 08.01. Comparison 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months), Outcome 01 pain (lower

values mean better) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

128Analysis 08.02. Comparison 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months), Outcome 02

Improvement (higher values are better) . . . . . . . . . . . . . . . . . . . . . . . . .

129Analysis 08.03. Comparison 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months), Outcome 03 functional

status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

129Analysis 08.04. Comparison 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months), Outcome 04 physical

examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

129Analysis 08.06. Comparison 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months), Outcome 06

improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iiiAcupuncture and dry-needling for low back pain (Review)


130Analysis 09.01. Comparison 09 dry-needling versus other intervention ((Sub)acute LBP < 3 months), Outcome 01

global measure (higher values are better) . . . . . . . . . . . . . . . . . . . . . . . . .

131Analysis 09.02. Comparison 09 dry-needling versus other intervention ((Sub)acute LBP < 3 months), Outcome 02 Side

effects / Complications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

131Analysis 10.01. Comparison 10 acupuncture versus placebo or sham intervention (unknown / mixed duration of low

back pain), Outcome 01 pain (VAS): lower values are better . . . . . . . . . . . . . . . . . .


back pain), Outcome 02 global measure . . . . . . . . . . . . . . . . . . . . . . . . .


back pain), Outcome 03 functional status (higher scores are better). Generic instrument . . . . . . . .


back pain), Outcome 04 physical examination (higher values are better) . . . . . . . . . . . . . .

133Analysis 11.01. Comparison 11 acupuncture versus other intervention (unknown / mixed duration of low back pain),

Outcome 01 pain score (lower values mean better) . . . . . . . . . . . . . . . . . . . . .


Outcome 02 pain recovery: higher values are better . . . . . . . . . . . . . . . . . . . . .


Outcome 03 global measure (higher values are better) . . . . . . . . . . . . . . . . . . . .


Outcome 04 back specific functional status (higher scores are better). Ex: Japan Orthopedic Association Score.


Outcome 05 physical examination . . . . . . . . . . . . . . . . . . . . . . . . . . .


Outcome 07 Side effects / Complications . . . . . . . . . . . . . . . . . . . . . . . .

136Analysis 12.01. Comparison 12 acupuncture versus acupuncture. (unknown / mixed duration of low back pain),

Outcome 01 pain (lower values are better) . . . . . . . . . . . . . . . . . . . . . . . .


Outcome 02 pain recovery (higher values are better) . . . . . . . . . . . . . . . . . . . .


Outcome 03 global measure (higher values are better) . . . . . . . . . . . . . . . . . . . .


Outcome 04 functional status (higher values are better) . . . . . . . . . . . . . . . . . . .


Outcome 05 physical examination (finger-floor distance) Higher values are better. . . . . . . . . . .


Outcome 07 improvement . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

139Analysis 13.01. Comparison 13 acupuncture plus intervention versus other intervention alone. (Chronic LBP: > 3

months), Outcome 01 pain (lower values are better) . . . . . . . . . . . . . . . . . . . .


months), Outcome 02 pain: difference between within group changes . . . . . . . . . . . . . .


months), Outcome 03 pain disability index (lower values are better) . . . . . . . . . . . . . . .


months), Outcome 04 Pain: percentage of patients with >50% pain reduction . . . . . . . . . . .


months), Outcome 05 function: difference between within group changes . . . . . . . . . . . . .


months), Outcome 06 global measure . . . . . . . . . . . . . . . . . . . . . . . . .

144Analysis 13.07. Comparison 13 acupuncture plus intervention versus other intervention alone. (Chronic LBP: >

3 months), Outcome 07 back specific functional status (lower scores mean better). Ex: RDQ, Oswestry and

Aberdeen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .


months), Outcome 08 spine range of motion: difference between within group changes . . . . . . . .

ivAcupuncture and dry-needling for low back pain (Review)



months), Outcome 09 Side effects / Complications . . . . . . . . . . . . . . . . . . . . .

vAcupuncture and dry-needling for low back pain (Review)


Acupuncture and dry-needling for low back pain (Review)

Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L, Koes BW, Berman BM

This record should be cited as:

Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L, Koes BW, Berman BM. Acupuncture and dry-needling for low back

pain. Cochrane Database of Systematic Reviews 2005, Issue 1. Art. No.: CD001351. DOI: 10.1002/14651858.CD001351.pub2.

This version first published online: 24 January 2005 in Issue 1, 2005.

Date of most recent substantive amendment: 30 October 2004

A B S T R A C T

Background

Although low-back pain is usually a self-limiting and benign disease that tends to improve spontaneously over time, a large variety of

therapeutic interventions are available for its treatment.

Objectives

To assess the effects of acupuncture for the treatment of non-specific low-back pain and dry-needling for myofascial pain syndrome in

the low-back region.

Search strategy

We updated the searches from 1996 to February 2003 in CENTRAL, MEDLINE, and EMBASE. We also searched the Chinese

Cochrane Centre database of clinical trials and Japanese databases to February 2003.

Selection criteria

Randomized trials of acupuncture (that involves needling) for adults with non-specific (sub)acute or chronic low-back pain, or dry-

needling for myofascial pain syndrome in the low-back region.

Data collection and analysis

Two authors independently assessed methodological quality (using the criteria recommended by the Cochrane Back Review Group)

and extracted data. The trials were combined using meta-analyses methods or levels of evidence when the data reported did not allow

statistical pooling.

Main results

Thirty-five RCTs were included; 20 were published in English, seven in Japanese, five in Chinese and one each in Norwegian, Polish

and German. There were only three trials of acupuncture for acute low-back pain. They did not justify firm conclusions, because

of small sample sizes and low methodological quality of the studies. For chronic low-back pain there is evidence of pain relief and

functional improvement for acupuncture, compared to no treatment or sham therapy. These effects were only observed immediately

after the end of the sessions and at short-term follow-up. There is evidence that acupuncture, added to other conventional therapies,

relieves pain and improves function better than the conventional therapies alone. However, effects are only small. Dry-needling appears

to be a useful adjunct to other therapies for chronic low-back pain. No clear recommendations could be made about the most effective

acupuncture technique.

Authors’ conclusions

The data do not allow firm conclusions about the effectiveness of acupuncture for acute low-back pain. For chronic low-back pain,

acupuncture is more effective for pain relief and functional improvement than no treatment or sham treatment immediately after

treatment and in the short-term only. Acupuncture is not more effective than other conventional and “alternative” treatments. The

data suggest that acupuncture and dry-needling may be useful adjuncts to other therapies for chronic low-back pain. Because most of

the studies were of lower methodological quality, there certainly is a further need for higher quality trials in this area.



P L A I N L A N G U A G E S U M M A R Y

Thirty-five RCTs covering 2861 patients were included in this systematic review. There is insufficient evidence to make any recommen-

dations about acupuncture or dry-needling for acute low-back pain. For chronic low-back pain, results show that acupuncture is more

effective for pain relief than no treatment or sham treatment, in measurements taken up to three months. The results also show that for

chronic low-back pain, acupuncture is more effective for improving function than no treatment, in the short-term. Acupuncture is not

more effective than other conventional and “alternative” treatments. When acupuncture is added to other conventional therapies, it

relieves pain and improves function better than the conventional therapies alone. However, effects are only small. Dry-needling appears

to be a useful adjunct to other therapies for chronic low-back pain.

B A C K G R O U N D

Low-back pain is a major health problem among western indus-

trialized countries, and a major cause of medical expenses, absen-

teeism and disablement (van Tulder 1995). People with acute low-

back pain usually experience improvements in pain, disability, and

return to work within one month, further but smaller improve-

ments occur up to three months, after which, pain and disability

levels remain almost constant and most people will have at least

one recurrence within 12 months (Pengel 2003). Although low-

back pain is usually a self-limiting and benign disease (Waddell

1987), a large variety of therapeutic interventions are available to

treat it (van Tulder 1997). However, the effectiveness of most of

these interventions has not been convincingly demonstrated and

consequently, the therapeutic management of low-back pain varies

widely.

Acupuncture is one of the oldest forms of therapy and has its roots

in ancient Chinese philosophy. Traditional acupuncture is based

on a number of philosophical concepts, one of which postulates

that any manifestation of disease is considered a sign of imbal-

ance between the Yin and Yang forces within the body. In classical

acupuncture theory, it is believed that all disorders are reflected at

specific points, either on the skin surface or just below it. Vital

energy circulates throughout the body along the so-called meridi-

ans, which have either Yin or Yang characteristics. An appropriate

choice of the 361 classical acupuncture points located on these

meridians for needling is believed to restore the balance in the

body. When the needles have been placed successfully, the patient

is supposed to experience a sensation known as Teh Chi (in some

schools of traditional acupuncture). Teh Chi has been defined as

a subjective feeling of fullness, numbness, tingling, and warmth,

with some local soreness and a feeling of distension around the

acupuncture point. There is no consensus among acupuncturists

about the necessity of reaching Teh Chi for acupuncture to be

effective.

Since acupuncture disseminated to the west several hundred years

ago, many different styles of acupuncture have developed, in-

cluding Japanese Meridian Therapy, French Energetic Acupunc-

ture, Korean Constitutional Acupuncture and Lemington 5 Ele-

ment Acupuncture. While these are similar to traditional acupunc-

ture, they each have distinct characteristics. In recent decades,

new forms of acupuncture have developed, such as ear (auricular)

acupuncture, head (scalp) acupuncture, hand acupuncture and

foot acupuncture (Lao 1996). Modern acupuncturists use not only

traditional meridian acupuncture points, but also non-meridian

or extra-meridian acupuncture points, which are fixed points not

necessarily associated with meridians. Acupuncture commonly in-

cludes manual stimulation of the needles, but various adjuncts are

often used, including electrical acupuncture (in which an electri-

cal stimulator is connected to the acupuncture needle), injection

acupuncture (herbal extracts injected into acupuncture points),

heat lamps, and acupuncture with moxibustion (the moxa herb,

Artemisia vulgaris, is burned at the end of the needle) (Lao 1996).

Dry-needling is a technique that uses needles to treat myofascial

pain in any body part, including the low-back region. Myofas-

cial pain syndrome is a disease of muscle that produces local and

referred pain. It is characterized by a motor abnormality (a hard

band within the muscle) and by sensory abnormalities (tenderness

and referred pain). It is classified as a musculoskeletal pain syn-

drome that can be acute or chronic, regional or generalized. It can

be a primary disorder causing local or regional pain syndromes, or

a secondary disorder that occurs as a consequence of some other

condition (Gerwin 2001). In 1983, Travel and Simons published

the book Myofascial Pain and Dysfunction - the Trigger Point Man-ual (Travel and Simons 83), which shows the pain pattern of trig-

ger points in every muscle of the body. Myofascial trigger points,

once carefully identified, can be inactivated by various methods

including systemic muscle relaxants, botulinum toxin, antidepres-

sants, deep muscle massage (for example: Shiatsu), local injection

of substances such as steroids or lidocaine, and dry-needling. Dry-

needling involves the insertion of a needle (it can be an acupunc-

ture needle or any other injection needle without injecting any

liquid) at these trigger points. The needles are not left in situ, they

are removed once the trigger point is inactivated. The inactiva-

tion of the trigger point should be followed by exercises (usually

stretching) or ergonomic adjustments with the purpose to re-es-

tablish a painless, full range of motion, and avoid recurrences.

It is still unclear what exact mechanisms underlying the action of

acupuncture or dry-needling. Western scientific research has pro-

posed mechanisms for the effect of acupuncture on pain relief. It



has been suggested that acupuncture might act by principles of

the gate control theory of pain. One type of sensory input (low-

back pain) could be inhibited in the central nervous system by an-

other type of input (needling). Another theory, the diffuse noxious

inhibitory control (DNIC), implies that noxious stimulation of

heterotopic body areas modulates the pain sensation originating

in areas where a subject feels pain. There is also some evidence

that acupuncture may stimulate the production of endorphins,

serotonin and acetylcholine within the central nervous system, en-

hancing analgesia (Chu 1979; Stux 2003).

The effectiveness of acupuncture in the treatment of low-back

pain has been systematically reviewed before (van Tulder 1999

(a); van Tulder 1999 (b)) with inconclusive results due to the low

methodological quality of the included studies. This is an updated

review of all available scientific evidence, including evidence from

Chinese and Japanese trials, on the effectiveness of acupuncture

for both acute and chronic low-back pain, and dry-needling for

myofascial pain syndrome in the low-back region.

O B J E C T I V E S

The objectives of this systematic review were to determine the

effects of acupuncture for (sub)acute and chronic non-specific low-

back pain, and dry-needling for myofascial pain syndrome in the

low-back region, compared to no treatment, sham therapies, other

therapies, and the addition of acupuncture to other therapies.

C R I T E R I A F O R C O N S I D E R I N G

S T U D I E S F O R T H I S R E V I E W

Types of studies

Only randomized controlled trials (RCTs), with no language re-

striction, were included in this systematic review.

Types of participants

Adults (>18 years) with non-specific low-back pain and myofascial

pain syndrome in the low-back region were included. RCTs that

included subjects with low-back pain caused by specific patholog-

ical entities such as infection, metastatic diseases, neoplasm, os-

teoarthritis, rheumatoid arthritis or fractures were excluded. Low-

back pain associated with sciatica as the major symptom, preg-

nancy and post-partum were also excluded. Although some studies

did not exclusively limit the study population to patients with non-

specific symptoms, studies were included if the majority of the pa-

tients had non-specific low-back pain according to the predefined

criteria. Patients with (sub)acute (12 weeks or less) or chronic low-

back pain (more than 12 weeks), were included.

Types of intervention

Articles evaluating acupuncture or dry-needling treatments that

involve needling were included in this review. Acupuncture was

defined as “the diagnosis was made using traditional acupuncture

theory and the needles were inserted in classical meridian points,

extra points or ah-shi points (painful points)’. Dry-needling was

defined as ”the cause of pain was diagnosed as “Myofascial Pain

Syndrome”, the points were chosen by palpation in the muscle,

and the needles were inserted into these myofascial trigger points’.

Studies were included regardless of the source of stimulation (e.g.,

hand or electrical stimulation). Studies in which the acupuncture

treatment did not involve needling, such as acupressure or laser

acupuncture were excluded. The control interventions were no

treatment, placebo/sham acupuncture or other sham procedure,

and other therapeutic interventions. Trials comparing two tech-

niques of acupuncture or dry-needling were included, but anal-

ysed separately.

Types of outcome measures

RCTs were included that used at least one of the four outcome

measures considered to be important in the field of low-back pain:

pain intensity (eg, visual analog scale (VAS)), a global measure (eg,

overall improvement, proportion of patients recovered, subjective

improvement of symptoms), back specific functional status (eg,

Roland Disability Scale, Oswestry Scale) and return to work (eg,

return to work status, number of days off work). The primary

outcomes for this review were pain and functional status. Physi-

ological outcomes of physical examination (eg, range of motion,

spinal flexibility, degrees of straight leg raising or muscle strength),

generic health status (eg, SF-36, Nottingham Health Profile, Sick-

ness Impact Profile) and other symptoms, such as medication use

and side effects were considered secondary outcomes.

S E A R C H M E T H O D S F O R

I D E N T I F I C A T I O N O F S T U D I E S

See: Cochrane Back Group methods used in reviews.

The previous review had searched the literature from 1966 until

1996. The following search strategies were used for this updated

review:

1. CENTRAL, The Cochrane Library 2003, Issue 1;

2. MEDLINE (OVID) from 1996 to February 2003 (see Table

05 for strategy);

3. EMBASE (OVID) from 1996 to February 2003 (see Table 05

for strategy);

4. The Cochrane Back Review Group Trials Registry;

5. The Chinese Cochrane Centre Trials Registry;

6. A database search of controlled clinical trials published

in Japan, using “Igaku Chuo Zasshi” (Japana Centra Revuo

Medicina) web version (between 1987 - 2003);

7. Reference lists in review articles and trials retrieved;

8. Personal communication with experts in the field.



M E T H O D S O F T H E R E V I E W

Study selection

For this updated review, one author (ADF) generated the electronic

search strategies in CENTRAL, MEDLINE, and EMBASE and

downloaded the citations into Reference Manager 9.0 Two authors

(MvT and BK) then independently reviewed the information to

identify trials that could potentially meet the inclusion criteria.

Full articles describing these trials were obtained and the same

two authors independently applied the selection criteria to the

studies. Consensus was used to solve disagreements concerning

the final inclusion of RCTs and a third author was consulted if

disagreements persisted. One author (HT) searched and selected

the studies from the Japanese databases. The Chinese Cochrane

Centre generated the searches in their Trials Register and one

author (LXL) selected the studies. The authors of recent original

studies were contacted to obtain more information when needed.

Methodological quality assessment

The methodological quality of each RCT was independently

assessed by two authors (not always the same pair of authors).

Review authors were not blinded with respect to authors,

institution and journal because they were familiar with the

literature. Consensus was used to resolve disagreements and a third

author was consulted if disagreements persisted.

The methodological quality of the RCTs was assessed by using

the criteria list recommended in the Updated Method Guidelines

for systematic reviews in the Cochrane Back Review Group (van

Tulder 2003) (Table 06). Each item was scored as “yes”, “no” or

“don’t know” according to the definitions of the criteria (Table

06).

The methodological quality assessment of the studies was used

for two purposes: First, to exclude studies with fatal flaws (such

as drop-out rate higher than 50%, statistically significant and

clinically important baseline differences that were not accounted

in the analyses). Studies that passed the first screening for fatal

flaws were classified into lower or higher quality: Higher quality

was defined as a trial fulfilling six or more of the 11 methodological

quality criteria and not having a fatal flaw. Lower quality trials

were defined as fulfilling fewer than six criteria and not having a

fatal flaw. The classification into higher/lower quality was used to

grade the strength of the evidence.

Data extraction

Two authors independently extracted the data on the study

characteristics, funding, ethics, study population, interventions,

analyses and outcomes. The authors of recent studies (published

in the past five years) were contacted to obtain more information

when needed.

Adequacy of treatment

Three authors, who are experienced acupuncturists (AF, LXL and

HT), judged the adequacy of treatment. The data extraction

included four questions about the adequacy of treatment,

which were derived from the STRICTA recommendations

(MacPherson 2002): 1) Choice of acupoints, 2) Number of

sessions, 3) Needling technique and 4) Acupuncturist experience.

The control groups were also judged as 1) appropriateness

of sham/placebo intervention and 2) adequate number of

sessions/dose. In addition, a panel of experts in acupuncture

treatment for low-back pain was consulted in a three-hour session

in which each study was presented for discussion (only the

population and interventions were presented, so the panel was

blinded to authors, journal, year, country, outcomes and results).

The panel consisted of six physicians trained in a variety of

acupuncture methods (Traditional Chinese medicine, Ryodoraku,

dry-needling, trigger point injections and scalp needling) who

work at a multidisciplinary pain clinic in Sao Paulo, Brazil. The

panel also classified each study as acupuncture or dry-needling.

Clinical Relevance

The two authors who extracted the data also judged the clinical

relevance of each trial using the five questions recommended

by Shekelle et al (Shekelle 1994) and the Updated Method

Guidelines(van Tulder 2003):

1. Are the patients described in detail so that you can decide

whether they are comparable to those that you see in your practice?

2. Are the interventions and treatment settings described well

enough so that you can provide the same for your patients?

3. Were all clinically relevant outcomes measured and reported?

4. Is the size of the effect clinically important?

5. Are the likely treatment benefits worth the potential harms?

Analysis

The primary analyses, decided a priori, were:

• acupuncture compared to no treatment, placebo or sham

therapy

• acupuncture compared to another intervention

• acupuncture added to an intervention compared to the

intervention without acupuncture.

Any other comparisons were considered secondary analysis.

The results of each RCT were plotted as point estimates, ie, relative

risks (RR) with corresponding 95% confidence interval (95% CI)

for dichotomous outcomes, mean and standard deviation (SD)

for continuous outcomes, or other data types as reported by the

authors of the studies. When the results could not be plotted,

they were described in the table of included studies or the data

were entered into “other data tables”. For continuous measures,

preference was given to analyse the results with weighted mean

differences (WMD) because these results are easier to interpret

for clinicians and other readers. If this was not possible, then

standardized mean differences (SMD) or effect sizes were used.

The studies were first assessed for clinical homogeneity with respect

to the duration of the disorder, types of acupuncture, control

group and the outcomes. Clinically heterogeneous studies were



not combined in the analysis, but separately described. For studies

judged as clinically homogeneous, statistical heterogeneity was

tested by Q test (chi-square) and I2. Clinically and statistically

homogeneous studies were pooled using the fixed effect model.

Clinically homogeneous and statistically heterogeneous studies

were pooled using the random effects model. Funnel plots were

constructed when at least 10 studies were available for the meta-

analysis (Sutton 2000).

When the data could not be entered in the meta-analysis because

of the way the authors of the trials reported the results (for

example: no information about standard deviation of the means)

we performed a qualitative analysis by attributing various levels of

evidence to the effectiveness of acupuncture, taking into account

the methodological quality and the outcome of the original studies

(van Tulder 2003):

• Strong evidence*-consistent** findings among multiple higher

quality RCTs

• Moderate evidence-consistent findings among multiple lower

quality RCTs and/or one higher quality RCT

• Limited evidence-one lower quality RCT

• Conflicting evidence-inconsistent findings among multiple trials

(RCTs)

• No evidence-no RCTs

* There is consensus among the Editorial Board of the Back Review

Group that strong evidence can only be provided by multiple

higher quality trials that replicate findings of other researchers in

other settings.

** When >75% of the trials report the same findings.

The results were grouped according to the following study

characteristics:

1) Type of acupuncture: two subgroups were analysed separately:

a. acupuncture in which the points were chosen by the meridian

theory

b. dry-needling in which needles were inserted in trigger points

2) Duration of pain: three subgroups were analysed separately:

a. acute and subacute pain (duration 12 weeks or less)

b. chronic (duration more than 12 weeks)

c. unknown or mixed duration

3) Control group:

a. no treatment

b. placebo or sham acupuncture

c. other interventions or acupuncture in addition to other

interventions

d. two different techniques of acupuncture

4) Outcome measures:a. Pain

b. Global measure

c. Functional status

d. Physical examination

e. Return to work

f. Complications

5) Timing of follow-up:

a. immediately after the end of the sessions - up to one week after

the end of the sessions

b. short-term follow-up - between one week and three months

after the end of the sessions

c. intermediate-term follow-up - between three months and one

year after the end of the sessions

d. long-term follow-up - one year or longer after the end of the

sessions

D E S C R I P T I O N O F S T U D I E S

The review published in 1999 included 11 studies (van Tulder

1999 (a)). This updated review includes 35 studies and 2861 pa-

tients. Twenty were published in English, seven in Japanese (Araki

2001; Inoue 2000; Inoue 2001; Kurosu 1979(a); Sakai 1998; Sakai

2001; Takeda & Nabeta 2001), five in Chinese,(Ding 1998; He

1997; Li & Shang 1997; Wang 1996; Wu (a) 1991), one in Nor-

wegian (Kittang 2001), one in Polish (Lopacz & Gralewski), and

one in German (Von Mencke 1988). The majority of the pop-

ulation included in these trials had chronic low-back pain (24

studies, 1718 patients). The control groups were the following:

no treatment, sham acupuncture, sham transcutaneous electrical

nerve stimulation (TENS), Chinese herbal medicine, education,

exercise, massage, moxibustion, non-steroidal anti-inflammatory

drugs, physiotherapy, spinal manipulation, TENS, trigger point

injections, and usual treatment by a general practitioner. Six stud-

ies compared the effectiveness of two different acupuncture tech-

niques.

M E T H O D O L O G I C A L Q U A L I T Y

The results of the methodological quality assessment are shown

in Additional Table 01. There were two studies with fatal flaws:

Giles & Muller 1999 had a 52% dropout during treatment pe-

riod in the acupuncture group and Grant 1999 had clinically im-

portant differences in the main outcome measures at baseline.

Therefore, these two trials are not included in the analyses or used

to draw conclusions. Of the remaining 33 trials, 14 were judged

to be of higher (Araki 2001; Carlsson 2001; Ceccherelli 2002;

Cherkin 2001; Garvey 1989; Inoue 2000; Inoue 2001; Kittang

2001; Leibing 2002; Meng 2003; Sakai 2001; Molsberger 2002;

Tsukayama 2002; Yeung 2003) and 19 to be of lower methodolog-

ical quality (Coan 1980; Ding 1998; Edelist 1976; Giles & Muller

2003; Gunn 1980; He 1997; Kerr 2003; Kurosu 1979(a); Kurosu

1979(b); Li & Shang 1997; Lehmann 1986; Lopacz & Gralewski;

MacDonald 1983; Mendelson 1983; Sakai 1998; Takeda & Na-

beta 2001; Thomas & Lundberg 94; Von Mencke 1988; Wang

1996; Wu (a) 1991). In none of the 35 trials was the care provider



blinded; in 28 trials, the timing of the outcome assessment was

similar in all groups. The biggest problem was the quality of report-

ing, which did not allow us to judge the following items: method

of randomization (15 trials), concealment of allocation (16 tri-

als), baseline differences (18 trials), co-interventions (18 trials) and

compliance (17 trials). Of the seven trials published in Japanese,

four were of higher (Araki 2001; Inoue 2000; Inoue 2001; Sakai

2001) and three were of lower methodological quality. All five tri-

als published in Chinese were of lower methodological quality.

R E S U L T S

STUDY SELECTION

Our searches resulted in the identification of 68 in CENTRAL,

49 reports in MEDLINE, and 85 in EMBASE. We obtained hard

copies of 40 articles, but excluded 17 because they did not meet

our inclusion criteria. In addition, we retrieved 16 hard copies

of studies published in Japanese and 11 published in Chinese,

but excluded nine and six respectively, because they did not meet

our inclusion criteria. Reasons for the exclusion of these studies

are explained in the Table of Excluded Studies. We contacted the

primary authors of eight trials to obtain additional information

that was not reported in the published study. Six responded to our

requests - all from the Japanese language trials.

CLINICAL RELEVANCE

The results of the clinical relevance assessment of each included

study are shown in Table 02. It should be noted that there was an

enormous variance in the way the authors judged the five items of

clinical relevance. This occurred because different pairs of authors

assessed the 35 trials and each author has a different background

and training. In addition, there were no clear instructions of what

should constitute a “yes” or “no” response for each question. As a

consequence, the assessment of clinical relevance of each individual

trial is subjective and difficult to analyse in the context of this

systematic review. Additional Table 04 shows the improvement

in pain for each treatment group and for each duration of low-

back pain. The average improvement in pain with acupuncture

for acute low-back pain was 52% (based on two studies), 32% for

chronic (16 studies) and 51% for unknown or mixed durations of

pains (eight studies). The average improvement of pain with no

treatment was 6% (six studies). The average improvement of pain

with sham or placebo therapies was 22% for acute (one study),

23% for chronic (six studies) and 25% for unknown or mixed

durations of pain (three studies).

ADEQUACY OF ACUPUNCTURE

The results are shown in Table 03. In all trials, acupuncture was

judged to be adequate for the population they included.

PRIMARY ANALYSES

1. Acupuncture compared to no treatment, placebo or sham

therapy (See Figure 01)

1a. Acupuncture versus no treatment for acute low-back pain:

There is no evidence because we did not find any RCT for this

comparison.

1b. Acupuncture versus sham therapy for acute low-back pain:

We found only one RCT and it used only one session of bilateral

acupuncture on the SI3 acupoint. Therefore, there is moderate

evidence (one higher quality trial, 40 people) (Araki 2001) that

there is no difference in pain and function, between one session

of acupuncture on the SI3 acupoint bilaterally and sham needling

of the same point immediately after the session.

1c. Acupuncture versus no treatment for chronic low-back pain:

The pooled analysis of two lower quality trials (90 people) (Coan

1980; Thomas & Lundberg 94) shows that acupuncture is more

effective than no treatment for patients with chronic low-back

pain for short term pain relief, with a SMD of -0.73 (95% CI -

1.19 to -0.28) (See comparison 05.01). There is limited evidence

(one lower quality trial, 40 people) (Thomas & Lundberg 94)

that acupuncture is also more effective at intermediate follow-up

for outcomes of pain. The pooled analysis of two lower quality

trials (90 people) (Coan 1980; Thomas & Lundberg 94) shows

that acupuncture is more effective than no treatment for patients

with chronic low-back pain in short-term functional improve-

ment, with an effect size of 0.63 (95% CI 0.19 to 1.08) (compar-

ison 05.07). There is limited evidence (one lower quality trial, 40

people) (Thomas & Lundberg 94) that there is no difference at

the intermediate-term follow-up in functional outcome, between

acupuncture and no treatment.

1d. Acupuncture versus sham therapy for chronic low-back

pain: Six trials (three higher and three lower quality) measured

pain outcomes (Carlsson 2001; Kerr 2003; Lehmann 1986; Leib-

ing 2002; Mendelson 1983; Molsberger 2002), and one higher and

two lower quality trials measured functional outcomes (Lehmann

1986; Leibing 2002; Mendelson 1983). Of five trials that mea-

sured pain immediately after the end of the sessions, four tri-

als could be pooled (Mendelson 1983; Leibing 2002; Molsberger

2002; Kerr 2003). The pooled analysis (two higher and two lower

quality RCTs, 314 people) shows that acupuncture is more effec-

tive than sham therapy with a WMD of -10.21 (95% CI -14.99

to -5.44) (comparison 06.01). The trial not included in the meta-

analysis (Lehmann 1986) included 36 people and found a trend

that acupuncture was better than sham therapy, but failed to reach

statistical significance. This trial could not be pooled with the

other studies because of the scale they used to measure pain and

the way they analysed the results. For short-term measures of pain,

there is strong evidence (two higher quality trials, 138 people)

(Carlsson 2001; Molsberger 2002) that acupuncture is more ef-

fective than sham therapy for patients with chronic low-back pain,

with a WMD of -17.79 (95% CI -25.5 to -10.07) (See compari-

son 06.01 and other data table 06.09). There are three trials (two

higher and one lower quality, 255 people) that assessed intermedi-

ate-term pain (Carlsson 2001; Lehmann 1986; Leibing 2002). All



three trials found a trend that acupuncture was better than sham

therapy, but without statistical significance. It was possible to pool

two of these studies, showing a WMD of -5.74 (95% CI -14.72 to

3.25) (See comparison 06.01). The only exception was the analysis

adjusted for baseline values conducted by Carlsson and Sjolund

(See other data table 06.09.03) that showed a statistically signif-

icant effect (p=0.007) in favour of acupuncture over sham ther-

apy. For long-term measures of pain, there is moderate evidence

(one higher quality trial, 51 people) (Carlsson 2001) that there is

no difference between acupuncture and sham therapy for chronic

low-back pain. For measures of function taken immediately after

the end of the sessions, there is moderate evidence (one higher

and two lower quality trials, 316 people) (Lehmann 1986; Leib-

ing 2002; Mendelson 1983) that there is no difference between

acupuncture and sham therapy. For measures of function taken

at intermediate-term follow-up, there is moderate evidence (one

higher and one lower quality trials, 204 people) (Lehmann 1986;

Leibing 2002) that there is no difference between acupuncture

and sham therapy for patients with chronic low-back pain. There

is no evidence from RCTs on the effectiveness of acupuncture for

patients with chronic low-back pain for functional measures at

short or long-term follow-ups.

2. Acupuncture compared to another intervention (See Figure

02)

2a. Acupuncture versus other interventions for acute low-back

pain: There is moderate evidence (one higher quality trial, 57

people) (Kittang 2001) that there is no difference immediately

after, at the short-term, or at the intermediate-term follow-ups

between acupuncture and Naproxen 500 mg, taken twice daily for

10 days, in measures of pain (VAS).

2b. Acupuncture versus other interventions for chronic low-

back pain: Compared to spinal manipulation, there is limited ev-

idence (one lower quality trial, 68 people) (Giles & Muller 2003)

that acupuncture is less effective for measures of pain and function

immediately after the end of the sessions. Compared to massage,

there is moderate evidence (one higher quality trial, 172 people)

(Cherkin 2001) that there is no difference immediately after the

sessions in pain between acupuncture and massage, but there is a

statistically significant difference in favour of massage at the long-

term follow-up. For measures of function, massage was statisti-

cally significantly more effective than acupuncture immediately

after the end of the sessions, but there was only a marginally sta-

tistically significant difference in favour of massage at the long-

term follow-up. However, differences in effect were only small

(moderate evidence). Compared to celecoxib, rofecoxib or parac-

etamol, there is limited evidence (one lower quality trial, 72 peo-

ple) (Giles & Muller 2003) that there is no difference immediately

after the end of the sessions in measures of pain and function.

There is conflicting evidence (two trials, 56 people) (Tsukayama

2002; Lehmann 1986) on the effectiveness of acupuncture com-

pared to TENS for patients with chronic low-back pain for pain

measured immediately after the end of the sessions: one higher

quality trial with a small sample size (Tsukayama 2002) found

a statistically significant difference in favour of acupuncture over

TENS, while one lower quality trial (Lehmann 1986) found no

difference. There is limited evidence (one lower quality trial, 36

people) (Lehmann 1986) that there is no difference at the interme-

diate-term follow-up in pain between acupuncture and TENS for

patients with chronic low-back pain. There is moderate evidence

(one higher and one lower quality trial, 56 people) (Tsukayama

2002; Lehmann 1986) that there is no difference immediately after

the end of the sessions in functional ability, between acupuncture

and TENS, and there is limited evidence that there is no differ-

ence at the intermediate-term follow-up (Lehmann 1986). Finally,

compared to self-care education, there is moderate evidence (one

higher quality trial, 184 people) (Cherkin 2001) that there is no

difference immediately after the end of the treatments and at the

long-term follow-up in pain and function, between acupuncture

and self-care education.

3. Acupuncture added to an intervention compared to the in-

tervention without acupuncture (See Figure 02)

3a. Addition of acupuncture to other interventions for acute

low-back pain: Only one lower quality trial (100 people) (He

1997) showed that there is limited evidence that the addition of

acupuncture and moxibustion to Chinese herbal medicine is more

effective than Chinese herbal medicine alone for a global measure

of pain and function at the long-term follow-up.

3b. Addition of acupuncture to other interventions for chronic

low-back pain: There are four higher-quality trials that assessed

the effects of acupuncture added to other therapies and com-

pared it to the other therapy alone (289 people) (Leibing 2002;

Meng 2003; Molsberger 2002; Yeung 2003). The other thera-

pies included: exercises, NSAIDs, aspirin, non-narcotic analgesic,

mud packs, infrared heat therapy, back care education, ergonomics

or behavioural modification. The pooled analysis (comparison

13.01) shows that the addition of acupuncture to other interven-

tions is more effective than the other intervention alone for pain,

measured immediately after the end of the sessions (four higher

quality trials, 289 people) with a SMD of -0.76 (95% CI -1.02 to-

0.5), at the short-term follow-up (three higher quality trials, 182

people) with a SMD of -1.1 (95% CI -1.62 to-0.58), and at the

intermediate-term follow-up (two higher quality trials, 115 peo-

ple) with a SMD of -0.76 (95% CI-1.14 to-0.38). These effects

were also observed for functional outcomes (comparison 13.07)

immediately after the end of the sessions (three higher quality tri-

als, 173 people) with a SMD of -0.95 (95% CI -1.27 to-0.63), at

the short-term follow-up with a SMD of -0.95 (95% CI -1.37 to-

0.54), and at the intermediate-term follow-up with a SMD of -

0.55 (95% CI-0.92 to-0.18).

SECONDARY ANALYSES

1. Other outcome measures



Other outcome measures were extracted for the purpose of com-

plementing the conclusions based on the primary outcome mea-

sures.

1a. Global measures of improvement: Measures of global im-

provement included multiple-choice categorical scales (eg, im-

proved - same - worse) or dichotomous options (eg, improved -

not improved). In the case of multiple-choice categorical scales,

we dichotomized the categories according to the principle of “im-

proved” and “not improved”. The number of patients improved

was divided by the total number of patients in that group (com-

parison 03.02, 05.02, 06.02 and 07.02). These results were in

agreement with the result of the primary analysis, therefore they

do not change the conclusions and will not be discussed in this

review.

1b. Measures of work status: Measures of work status were basi-

cally the number of people who returned or had not returned to

work at follow-up. The pooled analysis of the two trials (one higher

and one lower quality, 58 people) (Carlsson 2001; Lehmann 1986)

that compared acupuncture to sham for chronic low-back pain

patients failed to show a difference at the intermediate-term fol-

low-up (comparison 06.06). Compared to TENS, there was one

lower quality trial (Lehmann 1986) that showed no difference in

return-to-work at the intermediate-term follow-up.

1c. Measures of physical examination: Measures of physical ex-

amination basically included range of motion of the lumbar region

measured, for example, by the finger-floor distance or Schober

tests (Araki 2001; Kerr 2003; Kittang 2001; Lehmann 1986; Leib-

ing 2002; Molsberger 2002; Takeda & Nabeta 2001; Thomas &

Lundberg 94; Von Mencke 1988) and a composite outcome mea-

sure based on physical exam (Edelist 1976; Wu (a) 1991; Wang

1996). We compared the agreement between the results of physical

examination with the results of pain and function in the trials that

reported these data. There were 16 situations in which pain and

physical examination were measured (eg, same trial, same com-

parison group, same follow-up, etc). There was agreement in 13

situations and disagreement in three. There were nine situations in

which functional outcomes and physical examination were mea-

sured (eg, same trial, same comparison group, same follow-up,

etc). There were five agreements and four disagreements.

1d. Measures of complications: Only 14 trials reported any mea-

sure of complications or side-effects (Carlsson 2001; Cherkin

2001; Garvey 1989; Giles & Muller 1999; Grant 1999; Kerr 2003;

Kittang 2001; Lehmann 1986; Leibing 2002; Meng 2003; Mols-

berger 2002; Sakai 2001; Tsukayama 2002; Yeung 2003). The re-

sults for complications that happened during the treatment period

showed that for a total of 245 patients who received acupuncture,

there were only 13 minor complications (5%), while for 156 pa-

tients who received sham therapy, there were no complications

(0%). In the group of 205 patients that received other interven-

tions (eg, TENS, NSAIDs, etc), there were 21 reports of compli-

cations (10%). None of the complications were fatal or so serious

that hospitalization was required.

2. Other comparisons

2a. Efficacy and effectiveness of dry-needling at trigger and mo-

tor points: (See Figure 03) There is limited evidence (one lower

quality trial, 17 patients) that superficial needling (4 mm) inserted

at trigger points is better than placebo TENS (MacDonald 1983).

Two randomized trials compared dry-needling with other inter-

ventions. There is limited evidence (one lower quality trial, 56

people) (Gunn 1980) that a few sessions of dry-needling, added to

a regimen of physiotherapy, occupational therapy and industrial

assessments is better than the regimen alone immediately after, at

the short and the intermediate-term follow-ups. There is moder-

ate evidence (one higher quality trial, 34 people) (Garvey 1989)

that there is no difference in short term global improvement be-

tween one session of dry-needling and one session of trigger point

injection with lidocaine and steroid, one session of trigger point

injection with lidocaine only, or one session of cooling spray over

the trigger point area followed by acupressure.

2b. Comparison between different techniques of acupuncture:

(See Figure 04)

i. For acute low-back pain, one single session of bilateral needling

of SI3 is better than one single session of needling of Yaotongxue

(Extra 29, EX-UE 7) (one lower quality trial, 150 patients) (Wu

(a) 1991)

ii. For chronic low-back pain, deep stimulation (1.5 cm in the

muscle or in the trigger point) is better than superficial stimulation

(2 mm in the subcutaneous tissue) immediately after the sessions

and at the short-term follow-up (one higher quality trial, 42 pa-

tients) (Ceccherelli 2002)

iii. For chronic low-back pain, the ancient needling technique

is better than the regular needling technique at the short-term

follow-up (one lower quality trial, 54 patients) (Ding 1998)

iv. For chronic low-back pain, manual acupuncture has the same

effects as electroacupuncture, both at the short and long-term

follow-ups (one higher quality trial, 34 patients) (Carlsson 2001)

v. For low-back pain of any duration, distal point needling is no

different from local lumbar area needling for measures of pain,

function and range of motion (one lower quality trial, 20 patients)

(Takeda & Nabeta 2001)

vi. For low-back pain of any duration, needle retention for about

10 minutes is better than removal immediately after insertion (one

lower quality trial, 20 patients) (Kurosu 1979(b)

vii. For low-back pain of any duration, local needling plus cupping

is more effective than distal treatment plus electrical stimulation

(one lower quality trial, 492 patients) (Wang 1996)

viii. For low-back pain of any duration, manual acupuncture plus

cupping is better than manual acupuncture alone (one lower qual-

ity trial, 156 patients) (Li & Shang 1997)

In summary, the best technique of acupuncture is still to be de-

termined, but the available high quality randomized trials suggest



that the best technique of acupuncture for low-back pain includes

deep stimulation (1.5 cm) instead of superficial stimulation (2

mm) and it seems that electrostimulation does not add any benefit

to manual stimulation of the needles.

2c. Efficacy and effectiveness of acupuncture for mixed popu-

lations of acute/chronic low-back pain: There were a few trials

that did not specify the duration of the low-back pain or that

mixed acute with chronic patients (Inoue 2000; Inoue 2001; Von

Mencke 1988; Sakai 1998; Kurosu 1979(a)). These trials will not

be discussed because they do not change the conclusions of this

review.

D I S C U S S I O N

Thirty-five RCTs covering 2861 patients were included in this

systematic review. There were only three trials of acupuncture for

acute low-back pain that do not justify firm conclusions, because

of small sample sizes and low methodological quality of the studies.

There is some evidence that acupuncture may be better than no

treatment or sham treatment for chronic low-back pain. However,

most studies have not found acupuncture to be more effective than

other conventional treatments (eg, analgesics, NSAIDs, TENS

and self-care education) or “alternative” treatments (eg, massage

or spinal manipulation). The data suggest that both acupuncture

and dry-needling may be useful adjuncts to other therapies for

chronic low-back pain.

Although the conclusions showed some positive results of

acupuncture, the magnitude of the effects were generally small.

The average pain reduction (measured by continuous scales such

as the VAS) in the group that received acupuncture for chronic

low-back pain was 32% compared to 23% in those who received

sham therapies and 6% in those who received no treatment. Fur-

thermore, the terms used to express the strength of the evidence

(strong, moderate and limited), as is standard in many systematic

reviews, might be misinterpreted. These are relative terms and are

often used to apply to a small number of “higher” quality studies.

This may give the false impression that “strong” evidence means

“definite” evidence, but this may not be the case.

Although efforts were made to find all published RCTs, some rel-

evant trials might have been missed. Twenty of the 35 included

RCTs were published in English, seven in Japanese, five in Chinese

and one each in Norwegian, Polish and German. Although no

languages were excluded, the number of non-English journals in-

dexed in electronic databases such as MEDLINE and EMBASE is

limited. If additional trials are found, this review will be updated.

The methodologic quality of the included RCTs, although im-

proving over the past several years, was poor. There were two stud-

ies with fatal flaws, and 14 studies with higher and 19 studies with

lower methodological quality. The methodologic quality in the

current review was defined by the internal validity criteria, which

referred to characteristics of the study that might be related to

selection, performance, attrition, and detection bias. It seems rea-

sonable that in the authors’ qualitative synthesis, the best evidence

would be provided by the higher quality studies, which are less

likely to have biased results. Although the levels of evidence in this

review may be considered arbitrary, it seems unlikely that a differ-

ent rating system would have resulted in different conclusions.

The included studies were very heterogeneous in terms of popula-

tion included, type of acupuncture administered, control groups,

outcome measures, timing of follow-up, and presentation of data.

Therefore, very few meaningful meta-analyses could be performed

and it was difficult to reach conclusions for most types of treat-

ments.

The experience and training of the acupuncturists who gave the

treatments were mentioned in a few studies. Some studies used

a protocol of a fixed set of points for all patients while others

used a flexible protocol where the points were selected for each

individual. Both methods are considered to be valid and were

analyzed together in this systematic review.

No serious adverse events were reported in the trials included in

this review. The incidence of minor adverse events was 5% in the

patients submitted to acupuncture. In the literature, most of the

reports of serious adverse events related to acupuncture are de-

scribed as case reports. In the past years, various prospective stud-

ies were conducted, enabling the estimation of the true incidence

of minor and major adverse events.

Melchart and colleagues reported the biggest prospective study,

covering over 760,000 treatments delivered by 7,050 German

physicians over a 10-month period. They observed 6,936 minor

(incidence of 91 per 10,000 treatments) and five major adverse

reactions (6 per 1,000,000 treatments), which included: exacerba-

tion of depression (one case), acute hypertensive crisis (one case),

vasovagal reaction (one case), asthma attack with hypertension

and angina (one case) and two cases of pneumothorax (Melchart

2004).

The other prospective studies did not observe any major adverse

reactions. Yamashita and colleagues observed 65,482 treatments

delivered by 84 therapists over a six-year period in Japan. There

were 94 cases of minor adverse events, with an incidence of 14

per 10,000 treatments, but this incidence was estimated using

data from spontaneous reports of adverse event by the practi-

tioner (Yamashita 1999). In another similar study by Yamashita

and colleagues, they forced practitioners to detect and report every

acupuncture session, whether there were adverse reactions or not.

Then, different incident rates of adverse reaction were obtained. A

total of 391 patients were treated in 1,441 sessions, involving a to-

tal of 30,338 needle insertions. The incidence of recorded systemic

reactions in individual patients was: tiredness (8.2%); drowsiness

(2.8%); aggravation of pre-existing symptoms (2.8%); itching in

the punctured regions (1.0%); dizziness or vertigo (0.8%); feeling



of faintness or nausea during treatment (0.8%); headache (0.5%);

and chest pain (0.3%) (Yamashita 2000).

MacPherson and colleagues observed 34,407 treatments delivered

by 574 Traditional Chinese Acupuncturists in the UK, over a four-

week period. There were 43 minor adverse events (incidence of

12.5 per 10,000 treatments) (MacPherson 2001). White and col-

leagues observed 31,822 treatments delivered by 78 acupunctur-

ists (physicians and physiotherapists) in the UK, over a 21-month

period. There were 43 minor adverse reactions (incidence of 13.5

per 10,000 treatments) (White 2001). Odsberg and colleagues

observed 9,277 treatments delivered by 187 physiotherapists in

Sweden over a four-week period, and recorded 2,108 minor ad-

verse reactions (incidence of 2,272 per 10,000 treatments) (Ods-

berg 2001). Ernst and colleagues observed 3,535 treatments de-

livered by 29 acupuncturists in Germany over a 13-month period,

and recorded 402 minor adverse reactions (incidence of 1,100 per

10,000 treatments) (Ernst 2003).

The great variation in incidence of minor adverse events is prob-

ably due to different definitions of adverse reaction, research de-

signs, or styles of acupuncture in the various studies.

Because serious adverse events are rare, they continue to be re-

ported in the form of case reports. Recently published systematic

reviews of case reports showed that these serious complications

may include infections (human immunodeficiency virus, hepati-

tis, bacterial endocarditis) caused by non sterile needles, and fa-

tal tissue trauma (pneumothorax, cardiac tamponade, spinal cord

injury) (Cherkin 2003; Chung 2003; Yamashita 2001). Further-

more, we have little information about the safety of acupuncture

specifically for low-back pain. We need more information about

the safety of acupuncture that focuses on specific conditions.

A U T H O R S ’ C O N C L U S I O N S

Implications for practice

There were only three heterogeneous trials of acupuncture for

acute low-back pain. Therefore we could not reach convincing

conclusion and there is a need for future studies to make recom-

mendation in this area.

There is some evidence of the effects of acupuncture for chronic

low-back pain. Compared to no treatment, there is evidence for

pain relief and functional improvement for acupuncture at shorter-

term follow-ups. Compared to sham therapies, there is evidence

for pain relief at shorter-term follow-up, but these effects were

not maintained at the longer-term follow-ups, nor were they ob-

served for functional outcomes. Compared to other conventional

or “alternative” treatments, acupuncture is no better for measures

of pain and function. There is evidence that acupuncture, added

to other conventional therapies, relieves pain and improves func-

tion better than conventional therapies alone. According to these

results, acupuncture may be useful as either a unique therapy for

chronic low-back pain or as an adjunct therapy to other conven-

tional therapies. Although the conclusions show some positive re-

sults of acupuncture, the magnitude of the effects were generally

small.

Although dry-needling appears to be a useful adjunct to other

therapies for chronic low-back pain, no clear recommendations

can be made because of small sample sizes and low methodological

quality of the studies.

With respect to the different techniques of acupuncture, most

studies were either small, of lower methodological quality, or both,

therefore, no clear recommendation could be made.

Implications for research

Because most of the studies were of poor methodological qual-

ity, there certainly is a need for future higher-quality RCTs. Also,

because many trials were poorly reported, we recommend that

authors use the CONSORT statement as a model for reporting

RCTs (www.consort-statement.org) and use the STRICTA crite-

ria (MacPherson 2002) to report the interventions. Many trials

could not be included in the meta-analyses because of the way

the authors reported the results, therefore we suggest that publi-

cations of future trials report means with standard deviations for

continuous measures, or number of events and total patients ana-

lyzed for dichotomous measures. Future research should focus on

areas where there are few or no trials, for example, acupuncture

compared to no treatment, placebo or sham for acute low-back

pain. Future studies should also have larger sample sizes, use a

valid acupuncture treatment, and have both a short-term and a

long-term follow-up (for chronic pain). From the available high

quality trials included in this review, deep stimulation seems to

be the most promising acupuncture treatment. Future studies are

needed that evaluate superior features of acupuncture. We suggest

that publications of future trials report the proportion of subjects

who obtain a clinically important improvement in the groups be-

ing compared to facilitate a judgment about clinically important

differences between the groups. Although an evaluation of costs

was not the objective of this review, we suggest that future research

assesses cost-effectiveness of acupuncture compared to other treat-

ments.

F E E D B A C K

March 2005

Summary

Feedback 1: When assessing the outcome of acupuncture therapy

for the low back, what points were used? What I have observed

is there is an immediate proprioceptive effect with the patients

following the therapy. Where there is a mild paresis on clinical ex-

amination, what I think acupuncture does is to establish a recruit-

ment of those muscle fibres that are paretic due to whatever cause.



Possible TypeII fibres are activated. Therefore, any post assessment

should not necessarily just assess pain but should include propri-

oceptive assessments, motor function and coordinatice activties.

Feedback 2: Which acupoints were used? What were the classi-

cal symptoms of pain being modified? My understanding is that

whilst acupuncture modifies pain, in doing so the manifestations

of pain are being treated. In the outcome of the study you men-

tion function as one of those outcomes. What were the functional

factors and how were they measured?

I am interested in the inclusion and exclusion criteria for partic-

ipants in the study, were there any controls, that is, particpants

without low back pain?

Author’s reply

Response 1: The outcomes were assessed immediately after the

end of treatment, and at short, intermediate and long-term fol-

low-ups. Definitions of these time-lines are given in the review.

The outcomes of interest were patient-reported pain and function.

The authors of the systematic review did not include neurological

outcomes and neither did the trials. We don’t know if data were

collected on these items in the original studies and not included

in the published reports.

Response 2: I think some of the details you are looking for can

be found in the ’Table of Included Studies’. If they are not listed,

its because they were not included in the published report of the

primary study, but to be sure, you may wish to refer to some of

the primary studies if you had particular questions. The full text

outlines which studies were included in these comparisons: [i]

Acupuncture compared to no treatment, placebo or sham therapy

[ii] Acupuncture compared to another intervention [iii] Acupunc-

ture added to an intervention compared to the intervention with-

out acupuncture. The authors also outline other outcomes and

comparisons in the results section. The inclusion criteria only in-

cluded Individuals with back pain.

Different aspects of pain and the tool used to measure them would

have been addressed in different studies ... this will be in the Table

of Included Studies; ditto for functional outcomes and measure-

ment tools and participants of each study.

Contributors

Dr Henare R Broughton, Occupation Family Physician

Dr Andrea Furlan, review author

P O T E N T I A L C O N F L I C T O F

I N T E R E S T

Three coauthors of this review (DC, HT and LXL) are also authors

of some included trials. In order to avoid any conflict of interest,

they were not involved in the methodological quality assessment

or data extraction of their own study.

A C K N O W L E D G E M E N T S

We would like to thank Maoling Wei from the Chinese Cochrane

Centre for searching the Chinese databases. We are grateful to

Mrs Gunn Elisabeth Vist who extracted the data from the Nor-

wegian paper and Marcos Hsu and Hitoshi Yamashita who were

the second authors for the Chinese and Japanese papers respec-

tively. We would like to thank the panel of experts for their im-

portant contribution to this review: Satiko Imamura, Marta Ima-

mura, Wu Tu Hsing, Helena Kazyama, Chien Hsin Fen and Lil-

iana George. We are also grateful to all authors who replied to our

requests to obtain more information. We also would like to thank

Sheilah Hogg-Johnson and Joseph Beyene for their assistance with

the statistical analyses. Finally we would like to thank the editors

of the Cochrane Back Review group who provided constructive

comments and Vicki Pennick, co-ordinator of the Cochrane Back

Review Group for her assistance and amendments.

Brian Berman’s work on this review was partially funded by Grant

Number R24 AT001293 from the National Center for Comple-

mentary and Alternative Medicine (NCCAM). The contents of

this article are solely the responsibility of the authors and do not

necessarily represent the official views of the NCCAM, or the Na-

tional Institutes of Health.

S O U R C E S O F S U P P O R T

External sources of support

• National Center for Complementary and Alternative Medicine

USA

Internal sources of support

• Institute for Work & Health CANADA

• Erasmus MC, Department of General Practice NETHER-

LANDS

• Vrije Universiteit, EMGO Institute NETHERLANDS



R E F E R E N C E S

References to studies included in this review

Araki 2001 {published and unpublished data}

Araki S, Kawamura O, Mataka T, Fujioka H, et al.Randomized con-

trolled trial comparing the effect of manual acupuncture with sham

acupuncture for acute low back pain [RCT ni yoru kyusei yotsu-sho

ni taisuru shishin-gun to gishin-gun no tiryou koka]. Journal of the

Japan Society of Acupuncture and Moxibustion 2001;51(3):382.

Carlsson (even) {published data only}

Carlsson (morn) {published data only}

Carlsson 2001 {published data only}

Carlsson CP, Sjolund BH. Acupuncture for chronic low back pain:

a randomized placebo-controlled study with long-term follow-up.

Clin J Pain 2001;17(4):296–305. [MEDLINE: 915].

Ceccherelli 2002 {published data only}

Ceccherelli F, Rigoni MT, Gagliardi G, Ruzzante L. Comparison of

superficial and deep acupuncture in the treatment of lumbar myofas-

cial pain: a double-blind randomized controlled study. Clin J Pain

2002;18(3):149–153. [MEDLINE: 1045].

Cherkin 2001 {published data only}

Cherkin DC, Eisenberg D, Sherman KJ, Barlow W, Kaptchuk TJ,

Street J, et al.Randomized trial comparing traditional Chinese med-

ical acupuncture, therapeutic massage, and self-care education for

chronic low back pain. Archives of Internal Medicine 2001;161(8):

1081–1088. [MEDLINE: 886].

Kalauokalani D, Cherkin DC, Sherman KJ, Koepsell TD, Deyo RA.

Lessons from a trial of acupuncture and massage for low back pain:

patient expectations and treatment effects. Spine 2001;26(13):1418–

1424.

Cherkin 2001 (mass) {published data only}

Cherkin 2001 (sc) {published data only}

Coan 1980 {published data only}

Coan RM, Wong G, Ku SL, Chan YC, Wang L, Ozer FT, et al.The

acupuncture treatment of low back pain: a randomized controlled

study. Am J Chinese Med 1980;8:181–189.

Ding 1998 {published data only}

Ding YD. Fly-probing-acupoint manipulation as a main treatment

for lumbago. Shanghai Journal of Acupuncture and Moxibustion 1998;

17(5):25–26. [MEDLINE: 4].

Edelist 1976 {published data only}

Edelist G, Gross AE, Langer F. Treatment of low back pain with

acupuncture. Canad Anaesth Soc J 1976;23:303–306.

Garvey 1989 {published data only}

Garvey TA, Marks MR, Wiesel SW. A prospective, randomized, dou-

ble-blind evaluation of trigger-point injection therapy for low-back

pain. Spine 1989;14:962–964.

Garvey 1989 (lidoc) {published data only}

Garvey 1989 (spray) {published data only}

Garvey 1989(steroid) {published data only}

Giles & Muller 1999 {published data only}

Giles LG, Muller R. Chronic spinal pain syndromes: a clinical pilot

trial comparing acupuncture, a nonsteroidal anti-inflammatory drug,

and spinal manipulation. 1999;22(6):376–381. [MEDLINE: 807].

Giles & Muller 2003 {published data only}

Giles LGF, Muller R. Chronic spinal pain. A randomized clinical trial

comparing medication, acupuncture and spinal manipulation. Spine

2003;28(14):1490–1503.

Giles 1999 (manip) {published data only}

Giles 1999 (NSAID) {published data only}

Giles 2003 (manip) {published data only}

Giles 2003 (NSAID) {published data only}

Grant 1999 {published data only}

Grant DJ, Bishop-Miller J, Winchester DM, Anderson M, Faulkner

S. A randomized comparative trial of acupuncture versus transcuta-

neous electrical nerve stimulation for chronic back pain in the elderly.

Pain 1999;82(1):9–13. [MEDLINE: 1081].

Gunn 1980 {published data only}

Gunn CC, Milbrandt WE, Little AS, Mason KE. Dry needling of

muscle motor points for chronic low-back pain: a randomized clinical

trial with long-term follow-up. Spine 1980;5:279–291.

He 1997 {published data only}

He RY. Clinical Observation on Treatment of Lumbago due to Cold-

Dampness By Warm-Acupuncture Plus Chinese Medicine. Chi-

nese Acupuncture & Moxibustion 1997;17(5):279–80. [MEDLINE:

1753].

Inoue 2000 {published and unpublished data}

Inoue M, Kitakouji H, Ikeuchi R, Katayama K, Ochi H, et

al.Randomized controlled pilot study comparing acupuncture with

sham acupuncture for lumbago [Yotsu ni taisuru gishin wo mochiita

randamuka hikaku-shiken no kokoromi]. Journal of the Japan Soci-

ety of Acupuncture and Moxibustion 2000;50(2):356. [MEDLINE:

1771].

Inoue 2001 {published and unpublished data}

Inoue M, Kitakouji H, Ikeuchi R, Katayama K, Ochi H, et

al.Randomized controlled pilot study comparing manual acupunc-

ture with sham acupuncture for lumbago (2nd report) [Yotsu ni

taisuru gishin wo mochiita randamuka hikaku-shiken no kokoromi].

Journal of The Japan Society of Acupuncture and Moxibustion 2001;51

(3):412.

Kerr 2003 {published data only}

Kerr DP, Walsh DM, Baxter D. Acupuncture in the management of

chronic low back pain: a blinded randomized controlled trial. The

Clinical Journal of Pain 2003;19:364–370.

Kittang 2001 {published data only}

Kittang G, Melvaer T, Baerheim A. [Acupuncture contra antiphlogis-

tics in acute lumbago]. Tidsskr Nor Laegeforen 2001;121(10):1207–

1210. [MEDLINE: 1101].

Kurosu 1979(a) {published data only}

Kurosu Y. Acupuncture and Moxibustion for Lumbago (II) - Com-

parative Experiment of the Therapeutic Effectiveness of Acupunc-

ture and Garlic Moxibustion. The Journal of the Japan Acupuncture

& Moxibustion Association 1979;28(2):31–34. [MEDLINE: 1765].



Kurosu 1979(b) {published data only}

Lehmann 1986 {published data only}

Lehmann TR, Russell DW, Spratt KF. The impact of patients with

nonorganic physical findings on a controlled trial of transcutaneous

electrical nerve stimulation and electroacupuncture. Spine 1983;8:

625–634.

∗Lehmann TR, Russell DW, Spratt KF, Colby H, Liu YK, Fairchild

ML, Christensen S. Efficacy of electroacupuncture and TENS in the

rehabilitation of chronic low back pain patients. Pain 1986;26:277–

290.

Leibing 2002 {published data only}

Leibing E, Leonhardt U, Koster G, Goerlitz A, Rosenfeldt JA, Hilgers

R, et al.Acupuncture treatment of chronic low-back pain - a random-

ized, blinded, placebo-controlled trial with 9-month follow-up. Pain

2002;96(1-2):189–196. [MEDLINE: 1113].

Li & Shang 1997 {published data only}

Li Q, Shang WM. The effect of acupuncture plus cupping on 78

cases with lumbago. Hebei Chinese Traditional Medicine 1997;19(5):

28. [MEDLINE: 1754].

Lopacz & Gralewski {published data only}

Lopacz S, Gralewski Z. A trial of assessment of the results of acupunc-

ture or suggestion in the treatment of low back pain. Neur Neurochir

Pol 1979;8:405–409.

MacDonald 1983 {published data only}

MacDonald AJ, Macrae KD, Master BR, Rubin AP. Superficial

acupuncture in the relief of chronic low back pain. Ann Royal Coll

Surg Engl 1983;65:44–46.

Mendelson 1983 {published data only}

Mendelson G, Kidson MA, Loh ST, Scott DF, Selwood TS, Kranz

H. Acupuncture analgesia for chronic low back pain. Clin Exp Neurol

1978;15:182–185.

∗Mendelson G, Selwood TS, Kranz H, Loh TS, Kidson MA, Scott

DS. Acupuncture treatment of chronic back pain.: a double-blind

placebo-controlled trial. Am J Med 1983;74:49–55.

Meng 2003 {published data only}

Meng CF, Wang D, Ngeow J, Lao L, Peterson M, Paget S. Acupunc-

ture for chronic low back pain in older patients: a randomized, con-

trolled trial. Rheumatology 2003;42:1–10.

Molsberger 2002 {published data only}

Molsberger A, Winkler J, Schneider S, Mau J. Acupuncture and con-

ventional orthopedic pain treatment in the management of chronic

low back pain - a prospective randomised and controlled clinical trial.

ISSLS. 1998:87.

Molsberger AF, Mau J, Pawelec DB, Winkler J. Does acupuncture

improve the orthopedic management of chronic low back pain--a

randomized, blinded, controlled trial with 3 months follow up. Pain

2002;99(3):579–587. [MEDLINE: 1760].

Sakai 1998 {published and unpublished data}

Sakai T, Tsukayama H, Amagai H, Kawamoto M, Masuda K, et

al.Controlled trial on acupuncture for lumbago [Yotsu ni taisuru hari

no hikaku-taisyo-shiken]. Journal of the Japan Society of Acupuncture

and Moxibustion 1998;48(1):110. [MEDLINE: 1773].

Sakai 2001 {published data only}

Sakai T, Tsutani K, Tsukayama H, Nakamura T, Ikeuchi T,

Kawamoto M, et al.Multi-center randomized controlled trial of

acupuncture with electric stimulation and acupuncture-like transcu-

taneous electrical nerve stimulation for lumbago. Journal of the Japan

Society of Acupuncture and Moxibustion 2001;51:175–84.

Yamashita H. Are the Effects of Electro-Acupuncture on Low Back

Pain Equal to those of TENS?. Focus on Alternative and Complemen-

tary Therapies 2001;6(4):254–255. [MEDLINE: 1752].

Takeda & Nabeta 2001 {published and unpublished data}

Takeda H, Nabeta T. Randomized controlled trial comparing the

effect of distal point needling with local point needling for low back

pain [RCT ni yoru yotsu-sho ni taisuru enkakubu-sisin to kyokusho-

sisin no koka hikaku]. Journal of the Japan Society of Acupuncture and

Moxibustion 2001;51(3):411.

Thomas & Lundberg 94 {published data only}∗Thomas M, Lundberg T. Importance of modes of acupuncture in

the treatment of chronic nociceptive low back pain. Acta Anaesthesiol

Scand 1994;38:63–69.

Tsukayama 2002 {published data only}

Tsukayama H, Yamashita H, Amagai H, Tanno Y. Randomised con-

trolled trial comparing the effectiveness of electroacupuncture and

TENS for low back pain: a preliminary study for a pragmatic trial.

Acupuncture in Medicine 2002;20(4):175–180. [MEDLINE: 1758].

Von Mencke 1988 {published data only}∗Von Mencke M, Wieden TE, Hoppe M, Porschke W, Hoffmann

O, Herget HF. Akupunktur des Schulter-Arm-Syndroms und der

Lumbagie/Ischialgie - zwei prosepktive Doppelblind-Studien* (Teil

I). Akupunktur 1988;4:204–215.

Von Mencke M, Wieden TE, Hoppe M, Porschke W, Hoffmann

O, Herget HF. Akupunktur des Schulter-Arm-Syndroms und der

Lumbagie/Ischialgie - zwei prosepktive Doppelblind-Studien* (Teil

II). Akupunktur 1989;5:5–13.

Wang 1996 {published data only}

Wang JX. The effect of acupuncture on 492 cases with lumbago.

Shanghai Acupuncture Journal 1996;15(5):28. [MEDLINE: 1755].

Wu (a) 1991 {published data only}

Wu YC ea. Acupuncture for 150 cases of acute lumbago. Shanghai

Journal of Acupuncture and Moxibustion 1991;10(2):18–19. [MED-

LINE: 3].

Wu (b) 1991 {published data only}

Yeung 2003 {published data only}

Yeung CKN, Leung MCP, Chow DHK. The use of electro-acupunc-

ture in conjunction with exercise for the treatment of chronic low-

back pain. The journal of alternative and complementary medicine

2003;9(4):479–490.

References to studies excluded from this review

Cai 1996

Cai Gw ZDLL. Clinic research in treatment of acute sciatica by

needling YaoYangKuan (DU3) deep. Shangai J of Acupuncture 1996;

15(2):8–9.



Duplan 1983

Duplan B, Cabanel G, Piton JL, Grauer Jl, Phelip X. Acupuncture

et lombosciatique a la phase aiguë: etude en double aveugle de trente

cas. Sem Hop Paris 1983;59:3109–3114.

Fox and Melzack 1976

Fox EJ, Melzack R. Transcutaneous electrical stimulation and

acupuncture: comparison of treatment for low-back pain. Pain 1976;

2(141-148).

Franke 2000

Franke A, Gebauer S, Franke K, Brockow T. [Acupuncture massage

vs Swedish massage and individual exercise vs group exercise in low

back pain sufferers--a randomized controlled clinical trial in a 2 x 2

factorial design]. Forsch Komplementarmed Klass Naturheilkd 2000;7

(6):286–293. [MEDLINE: 958].

Fujinuki 1989

Fujinuki R. Yobu sekityukan kyosakusyo no shinkyu tiryo ni kansuru

kenkyu (2). The study of acupuncture & moxibustion for lumbar

spinal canal stenosis (2). Journal of the Japan Acupuncture and Moxi-

bustion 1989;48(11):6–13. [MEDLINE: 1769].

Galacchi 1981

Gallacchi G, Muller W, Plattner GR, Schnorrenberger CC.

[Akupunktur - und Laserstrahlbehandlung beim Zervikal - and Lum-

balsyndrom]. Schweiz Med Wschr 1981;111:1360–66.

Gallacchi 1983

Gallacchi G, Muller W. [Akupunktur - bringt sie etwas?]. Schweiz

Rundschau Med 1983;72(22):778–82.

Ghia 1976

Guia JN, Mao W, Toomey T, Gregg JM. Acupuncture and chronic

pain mechanisms. Pain 1976;2(285-99).

Hackett 1988

Hackett GI, Seddon D, Kaminski D. Electroacupuncture compared

with paracetamol for acute low back pain. Practitioner 1988;232:

163–164.

Ishimaru 1993

Ishimaru K, Shinohara S, Kitade T, Yhodo M. Clinical efficacy of

electrical heat acupuncture (First report): effect on low-back pain.

American Journal of Acupuncture 1993;21(1):13–18.

Junnila 1982

Junnila SYT. Acupuncture therapy for chronic pain. American Journal

of Acupuncture 1982;10(3):259–62.

Kinoshita 1965

Kinoshita H. Comparative observation in Goshin-ho and Hinaishin-

po. The Journal of the Japan Acupuncture & Moxibustion Association

1965;18(2):5–9. [MEDLINE: 1763].

Kinoshita 1971

Kinoshita H. Consideration of tonification and dispertion based

upon clinical experiment. The Journal of the Japan Acupuncture &

Moxibustion Association 1971;20(3):6–13. [MEDLINE: 1762].

Kinoshita 1981

Kinoshita H, Kinoshita N. Clinical Research in the Use of Paraneural

Acupuncture for Sciatica. The Journal of the Japan Acupuncture &

Moxibustion Association 1981;30(1):4–13. [MEDLINE: 1767].

Koike 1975

Koike Y. Quantity of stimulation in the treatment of lumbago.

Acupuncture treatment for lumbago. The Journal of the Japan

Acupuncture & Moxibustion Association 1975;24(3):8–13. [MED-

LINE: 1761].

Kuramoto 1977

Kuramoto S. A clinical study of the effects of electrical acupuncture

on protrusions of the intervertebral discs. The Journal of the Japan

Acupuncture & Moxibustion Association 1977;26(2):45–48. [MED-

LINE: 1764].

Laitinen 1976

Laitinen J. Acupuncture and transcutaneous electric stimulation in

the treatment of chronic sacrolumbalgia and ischialgia. American

Journal of Chinese Medicine 1976;4(2):169–175.

Li 1994

Li J, Chenard JR, Marchand S, Charest J, Lavignolle B. Points

d’acupuncture et zones-gachettes: réponse a la presson et résistance

cutanée chez des lombalgies chroniques. Rhumatologie 1994;46:11–

19. [MEDLINE: 437].

Megumi 1979

Megumi N. Acupuncture-moxibustion Therapy for the Lumbago

known as Colic. The Journal of the Japan Acupuncture & Moxibustion

Association 1979;28(2):35–44. [MEDLINE: 1766].

Ren 1996

Ren, Tian-Ming. Needling Taichong (Liv 3) and Mingmen (Du 4 or

GV 4) to treat lower back pain. Journal of Clinical Acupuncture 1996;

12(5-6):90.

Shinohara 2000

Shinohara S, Kitade K, Tanzawa S. Effect of acupuncture based on

Jingjin (channel sinews) theory for musculoskeletal conditions [Un-

doki-kei shojo ni keikin no gainen wo katsuyo-shita rinsyo-hoho to

sono koka]. Journal of the Japan Society of Acupuncture and Moxibus-

tion 2000;50(2):340. [MEDLINE: 1770].

Sodipo 1981

Sodipo JOA. Transcutaneous electrical nerve stimulation (TENS)

and acupuncture: comparison of therapy for low-back pain. Pain.

1981:S277.

Sugiyama 1984

Sugiyama N, Ito F, Takagi T. The effect of acupuncture and mobi-

lization on lumbago. Journal of the Japan Society of Acupuncture and

Moxibustion 1984;33(4):402–9. [MEDLINE: 1768].

Wang 1997

Wang RY. The effect of acupuncture with moxibustion or acupunc-

true with cupping on 167 cases with lumbago. Anhui Chinese Tradi-

tional Medicine Clinical Journal 1997;9:272–3. [MEDLINE: 1756].

Wang 2000

Wang RR, Tronnier V. Effect of acupuncture on pain management

in patients before and after lumbar disc protrusion surgery - a ran-

domized control study. American Journal of Chinese Medicine 2000;

28(1):25–33.

Wedenberg 2000

Wedenberg K. A prospective randomized study comparing acupunc-

ture with physiotherapy for low-back and pelvic pain in pregnancy.

Acta Obstet Gynecol Scand 2000;79:331–335.

Xingsheng 1998

Xingsheng C. Comparative study on acupuncture needling methods

for sciatica: routine needling vs point-to-point penetration and deep

puncture. American Journal of Acupuncture 1998;26(1):37–41.



Xu 1996

Xu L, Zhi-xiang Z, Xian-ming L, Guang-zhan L, Cheng-xuan Q.

Acupuncture plus massage versus massage alone in treating acute

lumbar sprain. International Journal of Clinical Acupuncture 1996;7

(3):365–67.

Yue 1978

Yue SJ. Acupuncture for chronic back and neck pain. Acupuncture &

Electro-Therapeut Res Int J 1978;3:323–24.

Zhang 1995

Zhang ZT, Zhang QZ. The effect of acupuncture plus chiropractic

on 57 cases with waist protrusion of the intervertebral disk. Neck pain

and Lumbago Journal 1995;16(2):97–8. [MEDLINE: 1757].

Zhang 1996

Zhang HP, Du SP, u LJ. [Observation on therapeutic effects of

110 cases with sciatica treated by electro-acupuncture deeply at

Yaoyangguan]. Chinese Acupuncture & Moxibustion 1996;16(8):19–

20. [MEDLINE: 2].

Zhi 1995

Zhi L, Jing S. Clinical comparison between scalp acupuncture com-

bined with a single body acupoint and body acupuncture alone for

the treatment of sciatica. American Journal of Acupuncture 1995;23

(4):305–7.

References to ongoing studies

Cherkin

Efficacy of Acupuncture for Chronic Low Back Pain. Ongoing

study Funding: National Center for Complementary and Alternative

Medicine (NCCAM).

GerAc

German Acupuncture Trials. Ongoing study Starting date of trial not

provided. Contact author for more information.

Harvard Med School

Physical CAM Therapies for Chronic Low Back Pain. Ongoing study

Funding: NIH.

Kong

Ongoing study Starting date of trial not provided. Contact author

for more information.

Munglani

Randomised controlled single-blinded trial of deep intra-muscular

stimulation in the treatment of chronic mechanical low back pain..

Ongoing study Starting date of trial not provided. Contact author

for more information.

Thomas

Longer term clinical and economic benefits of offering acupuncture

to patients with chronic low back pain.. Ongoing study Funding:

NHS.

Additional referencesCherkin 2003

Cherkin DC, Sherman KJ, Deyo RA, Shekelle PG. A review of the

evidence for the effectiveness, safety and costs of acupuncture, mas-

sage therapy and spinal manipulation for back pain. Ann Int Med

2003;138:898–906.

Chu 1979

Chu LSW, Yeh SDJ, Wood DD. Acupuncture manual: a western ap-

proach. New York: Marcel Dekker Inc.

Chung 2003

Chung A, Bui L, Mills E. Adverse effects of acupuncture. Which are

clinically significant?. Canadian Family Physician 2003;49:985–89.

Ernst & White 1997

Ernst E, White A. Life-threatening adverse reactions after acupunc-

ture? A systematic review. Pain 1997;71:123–126.

Ernst 2003

Ernst G, Strzyz H, Hagmeister H. Incidence of adverse effects during

acupuncture therapy- a mutlicentre survey. Complementary Therapies

in Medicine 2003;11:93–97.

Gerwin 2001

Gerwin RD. Classification, epidemiology, and natural history of my-

ofascial pain syndrome. Current Pain and Headache Reports 2001;5

(5):412–20.

Lao 1996

Lao L. Acupuncture techniques and devices. J Alternative Comple-

mentary Med 1996;2:23–5.

MacPherson 2001

MacPherson H, Thomas K, Walters S, Fitter M. The York acupunc-

ture safety study: prospective survey of 34 000 treatments by tradi-

tional acupuncturists. BMJ 2001;323(486-7).

MacPherson 2002

MacPherson H, White A, Cummings M, Jobst KA, Rose K, Niemt-

zow RC, STRICTA Group. Standards for Reporting Interventions in

Controlled Trials of Acupuncture: the STRICTA recommendations.

J Altern Complement Med 2002;8(1):85–9.

Melchart 2004

Melchart D, Weidenhammer W, Streng A, Reitmayr S, Hoppe A,

Ernst E, et al.Prospective investigation of adverse effects of acupunc-

ture in 97,733 patients. Arch Intern Med 2004;164:104–5.

Odsberg 2001

Odsberg A, Schill U, Haker E. Acupuncture treatment: side effects

and complications reported by Swedish physiotherapists. Comp Ther

Med 2001;9:17–20.

Pengel 2003

Pengel LHM, Herbert RD, Maher CG, Refshauge KM. Acute low-

back pain: systematic review of its prognosis. BMJ 2003;327(323-

327).

Shekelle 1994

Shekelle PG, Andersson G, Bombardier C, et al.A brief introduction

to the critical reading of the clinical literature. Spine 1994;19:2028S–

31S.

Stux 2003

Stux G, Berman B, Pomeranz B. Basics of acupuncture. 5th Edition.

Berlin Heidelberg: Springer-Verlag, 2003.

Sutton 2000

Sutton AJ, Duval SJ, Tweedie RL, Abrams KR, Jones DR. Empirical

assessment of effect of publication bias on meta-analyses. BMJ 2000

June 10;320(7249):1574–7.

Travel and Simons 83

Travell JG, Simons DG. Myofascial Pain and Dysfunction: The Trigger

Point Manual. Williams & Wilkins, 1983.

van Tulder 1995

van Tulder MW, Koes BW, Bouter LM. A cost-of-illness study of

back pain in the Netherlands. Pain 1995;62:233–40.



van Tulder 1997

van Tulder MW, Koes BW, Bouter LM. Conservative treatment of

acute and chronic nonspecific low back pain. A systematic review

of randomized controlled trials of the most common interventions.

Spine 1997;22(18):2128–56.

van Tulder 1999 (a)

van Tulder MW, Cherkin D, Berman B, Lao L, Koes BW. Acupunc-

ture for low back pain. Cochrane Database Systematic Reviews 1999,

Issue 2.Art. No.: CD001351. DOI:10.1002/14651858.CD001351.

van Tulder 1999 (b)

van Tulder MW, Cherkin DC, Berman B, Lao L, Koes BW. The

effectiveness of acupuncture in the management of acute and chronic

low back pain. A systematic review within the framework of the

Cochrane Collaboration Back Review Group. Spine 1999;24(11):

1113–23.

van Tulder 2003

van Tulder M, Furlan A, Bombardier C, Bouter L, The Editorial

Board of the Cochrane Collaboration Back Review Group. Updated

method guidelines for systematic reviews in the Cochrane Collabo-

ration Back Review Group. Spine 2003;28(12):1290–9.

Verhagen 1998

Verhagen AP, De Vet HCW, De Bie RA, Kessels AGH, Boers M,

Knipschild PG. Balneotherapy and quality assessment: interobserver

reliability of the Maastricht criteria list and the need for blinded

quality assessment. Clin Epidemiol 1998;51:335–341.

Waddell 1987

Waddell G. A new clinical model for the treatment of low back pain.

Spine 1987;12:632–644.

White 2001

White A, Hayhoe S, Hart A, Ernst E. Adverse events following

acupuncture: prospective survey of 32 000 consultations with doc-

tors and physiotherapists. BMJ 2001;323:485–6.

Yamashita 1999

Yamashita H, Tsukayama H, Tanno Y, Nishijo K. Adverse events in

acupunture and moxibustion treatment: a six-year survey at a national

clinic in Japan. J Altern Complement Med 1999;5(3):229–36.

Yamashita 2000

Yamashita H, Tsukayama H, Hori N, Kimura T, Tanno Y. Incidence

of adverse reactions associated with acupuncture. J Altern Complement

Med 2000;6(4):345–50.

Yamashita 2001

Yamashita H, Tsukayama H, White AR, Tanno Y, Sugishita C, Ernst

E. Systematic review of adverse events following acupuncture: the

Japanese literature. Comp Ther Med 2001;9:98–104.

∗Indicates the major publication for the study

T A B L E S

Characteristics of included studies

Study Araki 2001

Methods -Randomized (draw lots). Used sealed opaque envelopes by the acupuncturist.

-Patients and outcome assessors were blinded.

-Funding: not reported

-Setting: private clinic in Osaka, Japan.

-Informed consent obtained orally from patients.

-Ethics approval: not described

-All patients were followed.

-Analysis: Mean difference between before and after. Repeated measure ANOVA for responses.

Participants 40 patients with acute low-back pain (less than three days) and no sciatica.

Diagnoses: lumbar disc herniation, discopathy and lumbago.

Mean age: 44 years old

28 males and 7 females.

Working status:?

Previous treatments:?

Co-morbidity:?

Interventions 1) The needles were inserted into SI3 (bilaterally) with Teh Chi sensation, in supine position, and then

patients were made to perform back exercise. Needles were left in situ during the back exercise. Insertion



ignorespaces http://dx.doi.org/10.1002/14651858.CD001351unskip unskip

Characteristics of included studies (Continued )

depth was 2.5 cm with stainless steel needles (50 mm length, 0.20 mm diameter). Acupuncture treatment

was performed once only.

Randomized to this group: 20

Acupuncturists’ experience: three and six years.

2) Sham needling was performed to SI3 (bilaterally) point in supine position. Acupuncturist mimicked

needle insertion: tapped head of needle guide tube and then patients were made to perform back exercise.

Gesture of needling was performed during the back exercise. Sham treatment was performed once only.


Outcomes 1) Pain: Visual Analog Scale (VAS) from 0 to 100 mm;

2) Function: Japan Orthopedic Association (JOA) score, ranges from 0 to 14 (higher is better). Used only

the category of restriction of daily activities.

3) Flexion: Finger-to-floor distance

All three outcomes were taken before and immediately after the single session.

Costs: not reported

Complications: not reported

Notes The original study was published in abstract only. We obtained additional information from the authors.

Language: Japanese

For results, see the comparisons:

02.01

02.03

02.04

02.06

02.07

02.08

Conclusion: “There is no difference between the effect of acupuncture and that of sham acupuncture”

Allocation concealment A – Adequate

Study Carlsson (even)

Methods see Carlsson 2001

Participants

Interventions

Outcomes

Notes

Allocation concealment D – Not used

Study Carlsson (morn)

Methods see Carlsson 2001

Participants

Interventions

Outcomes

Notes





Study Carlsson 2001

Methods - Randomized by computer generated list. A secretarial assistant who was not involved in the study performed

the assignments.

- Patients and outcome assessors were blinded.

- Funding: One author is supported by Swedish Medical Research Council.

- Setting: Pain clinic (outpatients) in Malmo General Hospital affiliated with University in Sweden.

- Informed consent: yes

- Ethics approval: yes

- Follow-up: 100% at one month, 62% at three months, and 53% at six months.

- Analysis: used “last observation carried forward” for missing values.

Baseline differences in pain (VAS) were resolved by analysing percent changes at follow-ups. However, for

this analysis they used the non-parametric Mann-Whitney test.

There is no information about which test they used to analyse the global assessments. But, when we replicate

the analysis using RevMan, we get different results from the authors if we use relative risks, but not if we use

odds ratios.

For sick leave they used Wilcoxon signed ranks test.

Participants 51 patients with low back pain for six months or longer (mean 9.5 years) without radiation below the knee

and normal neurological examination.

Diagnoses: 39 muscular origin, 11 severe structural changes on X-rays.

Excluded: trauma, systemic disease, pregnancy and history of acupuncture treatment.

Mean age: 50 years

17 males and 33 females.

Working status: 20 on sick leave, 17 retired, 12 full time, one unemployed.

Previous treatments: corsets, nerve blocks, analgesics, TENS, physiotherapy. Two had undergone surgery.

Interventions 1) Manual acupuncture: local points (BL24, BL25, BL26, Ex Jiaji) and distal points (LI11, LI4, BL40, BL57

and BL60). “Teh-Chi” feeling was sought in all instances, mostly at a needle-tip depth of 2 to 3 cm. The

needles were stimulated three times during the 20-minute treatment sessions to restore Teh Chi feelings. The

needles were disposable, stainless steel, with a diameter between 0.3 and 0.32 mm and a length between 30

and 70 mm.

Frequency: once per week for eight weeks; two further treatments were given during the follow-up assessments

period of six months or longer.

Randomized to this group:16

Acupuncturist’ experience: board certified anesthesiologist with more than 10,000 acupuncture treatments.

2) In addition to the needles as in the manual acupuncture group, they performed electrical stimulation of

four needles (one pair per side in the low back). Frequency: 2 Hz every 2.5 seconds, interrupted by a 15 Hz

train for 2.5 seconds.


3) Mock transcutaneous electrical nerve stimulation (TENS) given by an impressive, stationary, but dis-

connected GRASS (gradient-recalled acquisition in a steady state) stimulator attached to two large TENS

electrodes. The electrodes were placed on the skin over the most intensely painful area in the low back.

During stimulation, flashing lamps were displayed and visible to the patient. This group was seen once per

week for 8 weeks.


Outcomes 1) Pain: Visual Analog Scale (VAS) from 0 to 100 mm; measured in the morning and in the evening. Not

clear how many patients filled all pain diaries everyday.

2) Global assessment by physician. Subjective. Improvement is not defined

3) Present work status: number of people on sick leave.

4) Intake of analgesics recorded daily




5) Sleep quality recorded daily

Outcomes were taken at 1 month, 3 months and 6 months or longer after the end of the 8 sessions.

The results of these outcomes at baseline are not reported, except for pain which is slightly different between

acupuncture and placebo.

Costs: not reported

Complications: no complications occurred during treatment or follow-up period

Notes Language: English

Publication: full paper

Additional information from authors: no

The authors pooled groups 1 and 2 and compared with group 3.

The results for pain are similar in the morning and evening measurements.


06.01

06.02

06.06

06.08

06.09 (other data table)


08.02

Conclusion: “The authors demonstrated a long-term pain-relieving effect of needle acupuncture compared

with true placebo in some patients with low-back pain”


Study Ceccherelli 2002

Methods -Randomized (table of random numbers). No description of allocation concealment.

-Outcome assessors were blinded.

-Funding: AIRAS (Associazone Italiana per la Ricerca e l’Aggiornamento Scientifico)

-Setting: Pain clinic, University of Padova, Italy.

-Informed consent and ethics approval not reported

-All patients were followed

-Analysis: Between groups were initially compared by repeated measurements two-way ANOVA. Post hoc

comparison was done by the Bonferroni correction of the unpaired t-test.

Participants 42 patients with continuous pain for more than 3 months. Normal neurologic exam. No signs of radicular

compression.

Diagnoses: chronic lumbosacral myofascial pain.

Excluded: spinal cord injury, osteoporosis, rheumatic diseases, disk herniation, fibromyalgia, organic diseases,

hypertension or obesity.

Age: between 30-50 years old. Mean 42 years old.

30 males and 12 females

Working status: ?

Previous treatments: none had been treated with acupuncture

Co-morbidity: ?




Interventions 1) Deep acupuncture: 1.5 cm in the muscle or in the trigger point. Needles: disposable Sedatelec 300um

diameter of 3 different lengths: 10 mm, 29 mm and 49 mm. Points: Extra 19, VG6. The following were

inserted bilaterally: GB34, UB54, UB62. Plus four trigger points or as second choice in the four most painful

muscular tender points found in the lumbar area. Total of eight sessions (total 6 weeks), each session lasted

for 20 minutes.

All needles were stimulated for 1 minute immediately after the insertion and for 20 s. every 5 min at 5, 10

and 15 minutes. The frequency of alternate right and left rotation of the needles was 2 Hz.

Randomized to this group: not described

Acupuncturist’s experience: not described

2) Same as described for acupuncture, but the depth of insertion was only 2 mm in the skin.

Randomized to this group: not described

Outcomes 1) Pain: verbally using the McGill Pain Questionnaire. They used the number of words chosen and the pain

rating index. The pain rating index is the sum of numerical values that has been assigned to each word used

to describe the pain.

Measured immediately after the end of the sessions and after 3 months.

Costs: not reported




Additional information from authors: we contacted authors, but no response was received.


08.01

Conclusions: “Clinical results show that deep stimulation has a better analgesic effect when compared with

superficial stimulation”

Allocation concealment B – Unclear

Study Cherkin 2001

Methods -Randomized (computer-generated random sequence). A research assistant confirmed eligibility, collected

baseline data and randomized the eligible ones.


-Funding: Group Health Cooperative, The Group Health Foundation (Seattle), Wash and the John E. Fetzer

Institute (Kalamazoo) and Agency for Health Care Research and Quality, Rockville.

-Setting. Health Maintenance Organization in Washington State, USA

-Informed consent: yes

-Ethics approval: yes

-Follow-up: 95% at 4 weeks, 95% at 10 weeks and 95% at 52 weeks.

-Analysis: Intention-to-treat. ANCOVA for continous variables and Logistic regression for dichotomous

variables. Adjustments for baseline values: Roland score, baseline symptom bothersomeness scale score, pain

below the knee, more than 90 days of back pain, satisfaction with previous back care, sex and age.

Participants 262 patients who visited a primary care physician for low-back pain who had persistent pain for at least 6

weeks.

Diagnoses: Non-specific low-back pain.

Excluded: sciatica, acupuncture or massage for back pain, back care from a specialist or CAM provider, clotting

disorders or anticoagulant therapy, cardiac pacemakers, systemic or visceral disease, pregnancy, litigation or

compensation, inability to speak English, severe or progressive neurologic deficits, previous lumbar surgery,




recent vertebral fracture, serious comorbid conditions and bothersomeness of back pain less than 4 (on a 0

to10 scale).

Mean age: 44.9 years old

42% males and 58% females

Working status: 84% employed or self-employed

Treatments being received at the time of entry in the study: medications (68%), massage (16%), acupuncture

(3%), narcotics (10%)

Co-morbidity: see exclusion criteria

Interventions 1) Acupuncture: Traditional Chinese Medical acupuncture by licensed acupuncturists with at least 3 years of

experience; Basic TCM needling techniques, electrical stimulation and manual manipulation of the needles,

indirect moxibustion, infrared heat, cupping, and exercise recommendation.

Proscribed: massage including acupressure, herbs and treatments not considered common TCM (Japanese

meridian therapy). Number and location of needles were left to the provider. They were allowed up to 10

visits over 10 weeks for each patient. All patients were needled and “teh chi” was reported for 89%. Mean of

12 needles (range 5-16) were inserted in each visit. Acupuncturists recommended exercise for about half of

their patients, usually stretching, walking or swimming.

Randomized to this group: 94 (88 received acupuncture as randomized).

2) Massage by a licensed therapist with at least 3 years of experience. Manipulation of soft tissue: Swedish

(71%), movement reeducation (70%), deep-tissue (65%), neuromuscular (45%), and trigger and pressure

point (48%), moist heat or cold (51%). Prohibited: energy techniques (Reiki, therapeutic touch), meridian

therapies (acupressure and shiatsu) and approaches deemed too specialized (craniosacral and Rolfing). Massage

therapists recommended exercise. They were allowed up to 10 visits over 10 weeks per patient.

Randomized to this group: 78 (74 received massage as randomized).

3) Self-care education: high-quality and inexpensive educational material designed for persons with chronic

back pain: a book and 2 professionally produced videotapes.


Outcomes 1) Pain: bothersomeness of back pain (0 to10), leg pain (0 to10) or numbness or tingling (0 to10). The

higher score was used.

2) Function: Roland Disability Scale

3) Disability: National Health Interview Survey

4) Utilization: provider visits, X-rays, operations, hospitalizations, medication use, visits to other massage

therapists or acupuncturists

5) Costs

6) Satisfaction

7) SF-12 Mental and Physical Health summary scales

8) Number of days of exercise

Outcomes were measured at baseline, 4, 10 and 52 weeks after randomization

Complications: no serious adverse effects were reported by any study participant




07.01

07.03

07.08

However, the results shown in the table of comparisons are the unadjusted analysis. We based our conclusions

on the authors analyses.




Therefore, the results are presented in the other data table:

07.09

Conclusions: “Massage is an effective short-term treatment for chronic low-back pain, with benefits that

persist for at least one year. Self-care educational materials had little early effect, but by one year were almost

as effective as massage. If acupuncture has a positive effect, it seems to be concentrated during the first four

weeks because there was little improvement thereafter”.


Study Cherkin 2001 (mass)

Methods See Cherkin 2001

Participants

Interventions

Outcomes

Notes


Study Cherkin 2001 (sc)

Methods See Cherkin 2001

Participants

Interventions

Outcomes

Notes


Study Coan 1980

Methods -Randomization was carried out by having prepared in advance a small box with 50 identically-sized pieces

of paper, folded so that they could not be read. 25 had A and 25 had B written on them. The box was shaken

and one of the pieces of paper was removed from the box blindly.

-Nobody was blinded

-Funding: National Health and Medical Research Council of Australia

-Setting: Acupuncture Center in Maryland, USA

-Informed consent: ?

-Ethics approval:?


-Analysis: Adherers (or “per protocol analysis”).

Participants 50 patients recruited via newspapers with low-back pain for at least 6 months.

Diagnoses: Abnormal X-ray (38/43), Sciatica (27/49), Muscle spasm (36/46)

Inclusion criteria: no previous acupuncture treatments, no history of diabetes, infection or cancer, and not

more than 2 back surgeries.

Mean age: 47 years old (range 18 to 67)


Working status:?

Previous treatments: back surgery (4)

Interventions 1) Acupuncture: Classical Oriental meridian theory. Electrical acupuncture in some patients. Selection of

acupuncture loci varied. ’Acknowledged acupuncturists’. 10 or more sessions, approximately 10 weeks. Teh

chi unclear.





2) Waiting list, no treatment for 15 weeks. Then they received the same acupuncture treatment as above.


Outcomes 1) Pain: Mean pain scores (0=no pain and 10=worst pain)

2) Function: Mean limitation of activity (0=none and 3=severe)

3) Mean pain pills per week

4) Global improvement (improved, same, worse)

Results after 10 weeks in acupuncture and after 15 weeks in waiting list group

Costs: not reported


Notes The authors reported a per protocol analysis. However, because there is individual patient data reported in

the article, we were able to recalculate using the intention-to-treat principle.

Language: English


Additional information from authors: no


05.01

05.02

05.04

05.07

Conclusions: “This study demonstrated that acupuncture was a superior form of treatment for these people

with low-back pain, even though they had the condition for an average of 9 years”.


Study Ding 1998

Methods - Randomized (method not described). No mention of concealment of allocation.

-Patients blinded


-Setting: University in GuangZhou, China

-Informed consent: Not reported

-Ethics approval: Not reported


-Analysis: chi-squares between groups

Participants 54 patients with chronic low-back pain, frequent recurrence, worse during work and relief with rest.

Diagnosis: chronic low-back pain.

Excluded: specific pathological entities using lab tests and x-rays.

Mean age: 45 years old in the ancient needling technique and 42 in regular needling technique group (range

19-68)


Working status: ?

Previous treatments: ?

Interventions 1) Ancient needling technique “The turtle exploring the holes”. Major points: GV3, Ashi point(s). Supple-

ment points: BL40. Needles 0.38 mm X 75 mm were used for deeper insertion and to different direction




in 45 degree angle. Strong Teh chi sensation was obtained. The needles were retained for 40 to 50 minutes.

Treatments were given daily up to 10 treatments.


2) Regular needling technique. Needles 0.38 mm X 75 mm were used for deeper perpendicular insertion

with twirling or rotating technique was used until strong Teh Chi sensation was acquired. Needle retaining

was 20 minutes with 3 to 4 times twirling or rotating stimulation in between. Treatments were given daily

for up to 10 days.


Outcomes 1. Pain on a 4-point scale: “cure”: no pain for 2 months;

“marked effective”: pain markedly improved;

“improved”: pain is somewhat relieved; and “no change”.

Measured immediately after and 2 months after the end of the sessions.

Costs: Not reported

Complications: Not reported

Notes Language: Chinese


No additional information from authors

For results, see the other data table:

08.06

Conclusions: “An ancient needling technique is better than the regular needling technique in treating chronic

low back pain”.


Study Edelist 1976

Methods - Randomized (method not described). No mention of concealment of allocation.

-Outcome assessors blinded


-Setting: University Hospital in Toronto. Canada

-Informed consent: yes

-Ethics approval: yes

-Not sure if follow-up is complete

-Analysis: not reported

Participants 30 patients with low-back pain with no improvement after conventional therapy, including bed rest, anal-

gesics, heat and physiotherapy. Patients were felt to have disc disease, which could not be surgically improved.

Interventions 1) Acupuncture: Manual insertion of 4 sterile needles into traditional acupuncture points (BL 60 and BL

25 bilaterally) until reaching Teh Chi, then electroacupuncture at 3-10 Hz. 30 minutes, 3 treatments in

maximum 2 weeks. Training & experience of acupuncturists unknown.

Randomized to this group: not reported

2) Sham acupuncture, 4 needles placed in areas devoid of classic acupuncture points, no Teh Chi.

Randomized to this group: not reported

Outcomes 1) Global assessment: subjective improvement of back/leg pain

2) Global assessment: objective improvement as measured by increased range of spinal movement, improve-

ment in tests for nerve root tension and objective improvement in neurological signs.

Costs: not reported





Notes Number of patients randomized unknown. We only know that 30 were analysed.

We classified the patients into “chronic low-back pain”.

Language: English




06.02

06.05

Conclusions: “There seemed to be no difference in either the subjective or objective changes between the

two effects and suggest that much of the improvement in pain syndromes associated with acupuncture may

be on the basis of placebo effect”.

Allocation concealment C – Inadequate

Study Garvey 1989

Methods -Randomized (computer generated four-tier list). No mention of concealment of allocation.

-Patients and outcome assessors blinded. Therapists were blinded for content of injections (groups 2 and 3)


-Setting: Outpatient clinic in a hospital. USA

-Informed consent: not reported

-Ethics approval: not reported

-Follow-up: 51 of 63 randomized (81%)

-Analysis: Adherers and intention-to-treat (with worst case scenario). Continuity chi squared, adjusted test.

Participants 63 patients with acute non radiating low-back pain, normal neurological examination, absence of tension

signs, normal x-ray, persistent pain despite initial treatment of 4 weeks, being able to localize a point of

maximum tenderness (trigger point).

Age: mean 38 years old

Gender: 41 men and 22 women

Working status: not reported

Previous treatment: non-steroidal anti-inflammatory drugs, hot showers, avoidance of activities that aggravate

the pain. No exercise program had been started.

Interventions 1) Dry-needling stick with a 21-gauge needle after an isopropyl alcohol wipe. 1 session. Training & experience

of therapists unknown


2) injection with 1.5 ml of 1% lidocaine using a 1.5 inch, 21-gauge needle after an isopropyl alcohol wipe.


3) injection with 0.75 ml of 1% lidocaine and 0.75 ml of Aristospan (Triamcinolone Hexacetonide) using a

1.5 inch, 21-gauge needle after an isopropyl alcohol wipe.


4) 10-second ethyl chloride spray from 6 inches away, followed by 20 second acupressure using the plastic

needle guard after an isopropyl alcohol wipe.


Outcomes 1) global improvement: percentage of not improved or improved.

This outcome was measured at 2 weeks after the interventions.




Costs: Not reported

Complications:

Group 1) 1 case of “fever, chills and systemic upset”; 2 cases of increased pain due to intramuscular hematoma.

Group 3) “increased pain”

Notes Intervention is “dry-needling”

Language: English




09.01

09.02

Conclusions: “The injected substance apparently is not the critical factor, since direct mechanical stimulus

to the trigger-point seems to give symptomatic relief equal to that of treatment with various types of injected

medication”.


Study Garvey 1989 (lidoc)

Methods

Participants

Interventions

Outcomes

Notes


Study Garvey 1989 (spray)

Methods

Participants

Interventions

Outcomes

Notes


Study Garvey 1989(steroid)

Methods

Participants

Interventions

Outcomes

Notes





Study Giles & Muller 1999

Methods -Randomized (method not described). Person drew an envelope out of a box with 150 well-shuffled envelopes,

each containing one of three colour codes (50 envelopes per intervention)

-Outcome assessor and data analyst blinded

-Funding: Green Projects Donation fund Limited via the Royal Melbourne Institute of Technology and

partly supported by Townsville General Hospital and James Cook University

-Setting: Outpatient pain clinic in a hospital setting. Townsville Australia.

-Informed written consent was obtained.

-Ethical approval by the Northern Regional Health Authority’s Townsville General Hospital

-Follow up: 77 of 130 randomized (59%)

-Analysis: Based on “adherers only principle”, i.e.. discarded those who did not comply with the treatment

assigned. Checked for possible confounders and interactions by multiple regression and logistic regression.

Participants 77 patients with spinal pain for at least 13 weeks (median 6 years).

Diagnoses: 82% lower back pain; 42% neck pain and 34% upper back pain.

Excluded: nerve root involvement, spinal anomalities, pathology other than mild to moderate osteoarthrosis,

previous spinal surgery and leg inequality > 9mm.

Median age: 42 years old


Working status: 56% blue collar, 26% white collar, 13% academic, 5% retired

Previous treatments: 77% drugs, 42% manipulation, 40% physiotherapy and 6% acupuncture

Co-morbidity: not described

Interventions 1) The treating clinician decided which form of acupuncture to use.

One of four experienced medical acupuncturists using sterile HWATO Chinese disposable acupuncture

guide tube needles 50 mm long with a gauge of 0.25 mm for 20 minutes. An average number of 8 to 10

needles were placed in local tender points and in distant acupuncture points according to the “near and far”

technique, depending on the condition being treated. Once patients could satisfactorily tolerate the needles

for 20 minutes, low-volt electrical stimulation was applied to the needles. Six treatments were applied in a 3

to to 4-week.


Drop-outs: 26 (52%). Reasons: unrelated to the outcome

2) Spinal manipulation was performed as judged to be safe and appropriate by the treating chiropractor for

the spinal level of involvement only. A high-velocity, low-amplitude spinal manipulation was performed. Six

treatments applied in a 3 to to 4-week period.


Drop-outs: 13 (26%). Reasons: same as in the acupuncture group

3) Medication: tenoxican (20 mg/d) and ranitidine (50 mg x 2/ day). Medication was given to the patients

for the defined 3 to 4-week treatment period. Treatment times were standardized by arranging 15 to 20-

minute appointments for all visits to eliminate a potential placebo effect originating from different lengths

of exposure to the clinician


Drop-outs: 10 (33%). Reasons: same as in the acupuncture group.

Outcomes 1) Pain: Visual Analog Scale (VAS) from 0 to 10 cm

2) Pain frequency on 5-ordered categories: 1/month, 1/week, 1/day, frequent and constant.

3) Function: Oswestry Disability Index

4) Cross over to another intervention after the study period

All outcomes were measured immediately after the end of the treatment period




Costs: Not reported

Complications: No side effects occurred for acupuncture or manipulation. Three medically treated subjects

had gastric symptoms

Notes The results of this study are not used in this review because of the high drop-out rate in the acupuncture

group (52%) that might invalidate the results of this trial.

Language: English




07.01

07.03

07.08

Study conclusions: “the manipulation group displayed the most substantial improvements that were uni-

formly found to be significant. In the other intervention groups, not a single significant improvement could

be found in any of the outcome measures”.


Study Giles & Muller 2003

Methods -Randomized (method not described). Person drew an envelope out of a box with 150 well-shuffled envelopes,

each containing one of three colour codes (50 envelopes per intervention)

-Data analyst blinded

-Funding: Queensland State Government. Partly supported by Townsville General Hospital.

-Setting: Outpatient pain clinic in a hospital setting. Townsville Australia.

-Informed written consent was obtained.

-Ethical approval by the Northern Regional Health Authority’s Townsville General Hospital

-Follow up: 115 were randomized. Six dropped out before treatment for reasons not related to outcomes.

69/109 (63.3%) were followed.

-Analysis: Based on “intention-to-treat analysis”. Checked for possible confounders and interactions by

multiple regression and logistic regression.

Participants 109 patients with uncomplicated spinal pain for a minimum of 13 weeks (average duration was 6.4 years)

Diagnosis: mechanical spinal pain

Excluded: nerve root involvement, spinal anomalies, pathology other than mild to moderate osteoarthrosis,

spondylolisthesis exceeding grade 1, previous spinal surgery and leg length inequality >9 mm.

Median age: 39 years old


Working status: 29% skilled trade, 20% pensioner or unemployed, 20% manager, clerk or sales, 12%

professional, 18% other.

Previous treatments: not described

Co-morbidity: not described

Interventions 1) The clinician determined the form of acupuncture technique. The Near and Far technique consists of:

needling the trigger point and distal analgesia producing sympatholytic acupuncture points below the elbow

or knee.

Acupuncture was performed by one of two experienced acupuncturists using sterile disposable acupuncture

guide tube needles (length 50 mm, gauge 0.25 mm) during 20-minute appointments.




For each patient, 8 to 10 needles were placed in local paraspinal intramuscular maximum pain areas and

approximately 5 needles were placed in distal acupuncture point meridians depending on the spinal pain

syndrome being treated. Once patients could tolerate the needles, needle agitation was performed by turning

or “flicking” the needles at approximately 5-minute intervals for 20 minutes. The needles were inserted to a

length of 20 to 50 mm, in the maximum pain area, and up to approximately 5 mm in the distal points.

Two treatments per week up to the defined maximum of 9 weeks of treatment.

Randomized to this group: 36. Two were lost before treatment, 2 during treatment and 10 changed treatment

because of no effect.

2) Spinal manipulation. 20-minute appointment. High-velocity, low-amplitude thrust spinal manipulation

to a joint was performed as judged to be safe and usual treatment by the treating chiropractor for the spinal

level of involvement to mobilize the spinal joints. Two treatments per week up to a maximum of 9 weeks.

Randomized to this group: 36. One was lost before treatment, 1 during treatment and 8 changed treatments

because of “no effect”.

3) A medication could be selected that had not already been tried by a patient randomized into the mediation

arm of the study. The patients normally were given Celecoxib (200 to 400 mg/day) unless it had previously

been tried. The next drug of choice was Rofecoxib (12.5 to 25 mg/day) followed by paracetamol (up to 4

g/day). Doses, left to the sports physician’s discretion, were related particularly to the patient’s weight, with

the severity of symptoms playing a minor role. The treating sports physician also was allocated 20 minutes

for follow-up visits.

Randomized to this group: 43. Three were lost before treatment and 18 changed treatment (11 for “no effect”

and 8 for “side effects”)

Outcomes 1) Pain: Visual Analog Scale (VAS) from 0 to 10 cm

2) Pain frequency on 5-ordered categories: 1/month, 1/week, 1/day, frequent and constant.

3) Function: Oswestry Disability Index

4) Cross over to another intervention after the study period

5) SF-36 Health Survey Questionnaire

All outcomes were measured immediately after the end of the treatment period

Costs: Not reported

Complications: Not reported

Notes Not sure about proportion of patients with lower back pain.

The results might be biased by the high and differential drop out rates.

Results are presented as medians and 25th and 75th percentiles and were transformed to means and standard

deviations.

Language: English




07.01

07.03

Study results: “Manipulation yielded the best results over all the main outcome measures except the Neck

Disability Index, for which acupuncture achieved a better result than manipulation”. “All three therapies

showed positive response according to the SF-36 general health status questionnaire”

Conclusions: “In patients with chronic spinal pain, manipulation, if not contraindicated, results in greater

short-term improvement than acupuncture or medication”.





Study Giles 1999 (NSAID)

Methods See Giles and Muller 1999

Participants

Interventions

Outcomes

Notes


Study Giles 1999 (manip)


Participants

Interventions

Outcomes

Notes


Study Giles 2003 (NSAID)


Participants

Interventions

Outcomes

Notes


Study Giles 2003 (manip)


Participants

Interventions

Outcomes

Notes


Study Grant 1999

Methods - Random numbers were used (method not described) to generate a sequence of sealed envelopes containing

the treatment code, the next available envelope being opened on the patient’s entry into the trial.


-Funding: Grant from the Trustees of the Liberton Hospital Endowment Funds

-Setting: Outpatients clinic in the United Kingdom

-Informed consent: not reported

-The study was approved by the Lothian Research Ethics Committee

-Follow-up: 57 out of 60 randomized

-Analysis: No intention-to-treat. Mann-Whitney U-tests for between group differences.

Participants 60 patients aged 60 years or over, with a complaint of pain of at least 6 months duration.

Diagnoses: chronic low-back pain.

Excluded: treatment with anticoagulants, systemic corticosteroids, dementia, previous treatments with

acupuncture or TENS, cardiac pacemaker, other severe concomitant disease, inability of patient or therapist

to apply TENS machine.






Previous treatments: not reported

Interventions 1) Two sessions of manual acupuncture weekly for 4 weeks, i.e. eight sessions in total. The needles were of a

standard size (32 gauge, 1.5 inch length with guide tube). Points were chosen for the individual patient as in

routine clinical practice, only using points in the back. Six needles were used on average at each treatment

with a minimum of two and a maximum of eight. Treatment sessions lasted for 20 minutes.

Randomized to this group: 32. Two dropped out during the study. Reasons: influenza and dental problem.

2) TENS: Standard machine (TPN 200, Physio-Med-Services) using 50 Hz stimulation with the intensity

adjusted to suit the patient, again as a routine clinical practice. The patient was given her/his own machine

to use at home, and instructed to use it during the day as required for up to 30 minutes per session to a

maximum of 6 hours per day. She/he was also seen for 20 minutes, twice weekly, by the physiotherapist,

ensuring the same contact with him. At each visit, symptoms were reviewed, treatment discussed and the

optimum use of the TENS machine ensured.

Randomized to this group: 28. One dropped out due to acute depression.

Co-interventions: The patients were advised to continue existing medication but not to commence any new

analgesics or any additional physical treatments for the duration of the trial.

Outcomes 1) Pain: visual Analog scale (0 to 200 mm).

2) Pain subscale of the 38-item Nottingham Health Profile part 1.

3) Analgesics consumption

4) Spinal flexion

These outcomes were taken at baseline, 4 days and 3 months after last treatment session.

Costs: not reported

Complications: 3 acupuncture patients reported dizziness and 3 TENS patients developed skin reactions.

(Comparison 07.08)

Notes The two groups appear different at baseline with respect to the four outcome measures. Patients in the

acupuncture group have higher VAS and NHP pain scores, reduced spinal flexion and lower tablet consump-

tion compared to the TENS group.

Because the authors had not adjusted for baseline values, no conclusions can be made based on this study.

We could try to obtain raw data from authors and run ANCOVA, but the data is also skewed and transfor-

mation is not appropriate.

Results:

07.01

07.08

Language: English



Conclusions: “A 4-week course of either acupuncture or TENS had demonstrable benefits on subjective

measures of pain (VAS and NHP score) and allowed them to reduce their consumption of analgesic tablets.

The benefits of both treatments remained significant 3 months after completion, with a trend towards further

improvement in the acupuncture patients.”





Study Gunn 1980

Methods -Randomized (randomized blocks, blocks defined by age and operation status; the first subject from each

block was assigned to the acupuncture treatment.)

-No information about concealment of allocation

-Nobody was blinded

-Funding: Workers’ Compensation board of British Columbia

-Setting: Pain Clinic in Richmond, British Columbia, Canada

-Informed consent: Yes

-Ethics approval: not reported

-Follow-up: 56 (100%) at discharge, 53 (95%) at 12 weeks and 44 (78%) at time of writing.

-Analysis: Analysis of covariance. No intention-to-treat.

Participants 56 males with chronic low-back pain of at least 12 weeks duration, who had 8 weeks of a standard clinic

regimen.

Diagnoses: disc diseases, low-back strain, spondylitis, spondylolisthesis, radiculopathy, low-back contusion,

pseudoarthrosis, disc protrusion, prolapsed disc, lumbar disc syndrome, post-laminectomy syndrome, neu-

ropathy, sciatica, nerve root compression, facet sprain, musculo-ligamentous strain, compression fracture,

interspinous ligament strain,

Excluded: Psychosomatic backache. Females.

Mean age: 40.6 years old (range 20 to 62 years)

Working status: all off work.

Previous treatments: some had surgery.

Interventions 1) Dry-needling: Standard therapy (physiotherapy, remedial exercises, occupational therapy, industrial as-

sessment) plus dry-needling on muscle motor points (non-meridian), 3 to 5 cm needles, direction of the

needle perpendicular to the skin, mechanical stimulation by pecking and twirling, low voltage (9V) electrical

stimulation interrupted direct current or phasic current.

Maximum of 15 treatments (average 8), once or twice a week. Training & experience unknown.


2) Standard therapy only (physiotherapy, remedial exercises, occupational therapy, industrial assessment).


Outcomes 1) Global improvement:

0: no improvement. Still disabled. Unable to return to any form of employment

+: Some improvement. Some subjective discomfort. Able to return to lighter employment.

++: Good improvement. Slight subjective discomfort but able to return to work and function at pre-accident

employment (or equivalent).

+++: Total improvement. No subjective discomfort. Returned to previous (or equivalent) employment.

The above was measured after discharge, 12 weeks after discharge and at the time of writing of the paper.

(all these varied)

Costs: not reported


Notes Intervention is dry-needling.

We dichotomized at 0 versus +/++/+++.

Language: English







09.01

Conclusion: “The group that had been treated with needling was found to be clearly and significantly better

than the control group with regard to status at discharge, at 12 weeks, and at final follow-up”.


Study He 1997

Methods -Randomized (method not reported). No information about concealment of allocation

-Patients were blinded

-Funding: Not reported

-Setting: outpatient clinic in a hospital. University Centre in Sichuan Province, China

-Informed consent: Not reported

-Ethics approval: Not reported

-Follow-up: All 100 patients were followed.

-Analysis: Not reported

Participants 100 patients with low-back pain (5 days to 6 months duration), with limited range of motion, and symptoms

worse in cold and rainy weather.

Excluded: kidney or bone disease confirmed by urine test and X-ray.

Age range: 22 to 79 years old




Interventions 1) Manual acupuncture with moxibustion plus Chinese herbal medicine. Two groups of points: 1) GV 4,

BL 22 , Ashi-points. 2) BL23, GV 3 and Extra 9 (L3-L4). Moxibustion was used 2 to 3 times on the handle

of the needles and needles were retained for 30 minutes. Treatments were given daily up to 10 treatments.

Teh Chi sensation was obtained. Herbal formula was given daily.


Experience: unknown

2) Chinese herbal treatment alone.


Outcomes 1) Overall assessment that includes pain, physical function, sensitivity to weather change and return to work.

According to this measure, patients are classified into:

a) cured: no pain, return to normal life and work, remains normal at one-year follow-up;

b) marked effective: pain is generally gone, but still feels uncomfortable in cold and damp weather;

c) improved: pain is markedly relieved, still feels uncomfortable in cold and damp weather, but better than

pre-treatment

d) no changes: no significant change.

The overall assessment was measured one year after the end of the sessions.

Costs: not reported


Notes We classified the duration as acute/subacute.

We dichotomized at a/b/c versus d.

Language: Chinese







03.02

Conclusion: “Manual acupuncture with moxibustion plus Chinese herbal medicine is better (p<0.01) than

Chinese herbal medicine alone for treating low-back pain with cold and dampness, based on TCM diagnosis”.


Study Inoue 2000

Methods -Randomized (computer generated numbers). Allocation was done by a centralized office using the Internet.



-Setting: University hospital in Kyoto, Japan.

-Written informed consent was taken from patients

-The Ethics Committee approved this study.

-Follow-up: All 27 patients were followed (100%)

-Analysis: Mann-Whitney’s U test was used for between group analysis.

Participants 27 patients with low back pain of unknown duration who attended the outpatient acupuncture clinic.

Excluded: (1) neurological findings, pain or numbness in lower extremity; (2) malignancy, (3) infection or

inflammatory disease; (4) fracture; (5) lumbago due to urological problem, gynaecological problem, digestive

problem or cardio-vascular problem; (6) patients who cannot stop other conflicting or ongoing treatments;

(7) problem of general condition; (8) dementia; (9) pregnancy.


Gender: no information

Working status: no information

Previous treatments: no information.

Interventions 1) Real acupuncture: Two needling points were chosen bilaterally from lumbar area (i.e. 4 points in total):

BL52 and extra point (yao-yan: EX-B7). Needles were inserted to a depth of 20 mm, manipulated by sparrow

pecking method for 20 seconds, and then removed. One treatment session was performed.


Acupuncturist had more than 10 years of experience.

2) Sham acupuncture: The same two points were chosen. Acupuncturist mimicked needle insertions: tapped

head of needle guide tube, then gesture of needling was performed for 20 seconds. One session.


Outcomes 1) Pain: visual analog scale (VAS) at the most restricted action immediately after the single session.

Costs: not reported


Notes Language: Japanese

Publication: abstract

We obtained additional information from authors.


10.01

Conclusion: “There was no difference between real needling and sham needling”.





Study Inoue 2001

Methods -Randomized (computer generated numbers). Allocation was done by a centralized office using the Internet.



-Setting: University hospital in Kyoto, Japan.

-Written informed consent was obtained.

-The Ethics Committee approved this study.

-Follow-up: All 21 patients were followed (100%)

-Analysis: Mann-Whitney’s U test was used for between group analysis.

Participants 21 patients with low-back pain of unknown duration who attended the outpatient acupuncture clinic were

included.



problem or cardio-vascular problem; (6) patients who can not stop other conflicting or ongoing treatments;

(7) problem of general condition; (8) dementia; (9) pregnancy.


Gender: no information

Working status: no information

Previous treatments: no information.

Interventions 1) Real acupuncture: One needling point was chosen from lumbar area: most painful locus was detected.

Needles were inserted and sparrow-picking technique was performed for 20 seconds. One session.


Experience: not reported

2) Sham acupuncture: One needling point was chosen from lumbar area: most painful locus was detected,

same as real acupuncture group. Acupuncturist mimicked needle insertion: tapped head of needle guide tube,

then gesture of needling was performed for 20 seconds. One session.


Outcomes 1) Pain: visual analog scale (VAS) at the most restricted action immediately after the single session.

Costs: not reported



Publication: abstract



10.01

Conclusion: “Real needling is superior to sham needling”.





Study Kerr 2003

Methods -Randomized (computer generated numbers). No information about who performed allocation of patients.

-Patients and outcome assessors were blinded

-Funding: Department of Health and Social Services for Northern Ireland

-Setting: outpatient clinic in a hospital

-Signed a consent form and were verbally advised as to the nature of the intervention. Patients were informed

that they would receive one of 2 different forms of treatment being investigated

-Ethics permission was obtained from the University of Ulster’s Research Ethical Committee

-46 of 60 randomized patients (76%) finished the trial period and 40/60 (66.7%) were followed at 6 months.

-Analysis: Only those who completed the study (46/60). T-tests for paired and independent samples.

Participants 60 patients with chronic low-back pain (> 6 months) with or without leg pain and with no neurologic deficits.

Mean duration of pain was 75.8 months.

Excluded: age < 18 years old, pregnancy, underlying systemic disorder, rheumatoid arthritis, osteoarthritis of

the spine or cancer.





Interventions 1) Same set of acupoints for everyone, regardless of the distribution of their symptoms: Bl23, Bl25, GB

30, Bl40, Ki3 (all bilateral) and GV4. Eleven needles were used in each session (Seirin acupuncture needles

N8, 0.30 x 50 mm, c-type needle). The needles were inserted until Teh Chi was produced. Position: prone.

Duration: 30 minutes. Needles were manually rotated to produce Teh Chi initially and at 10 to 20 minute

intervals. Sessions: 6 sessions, over a 6-week period.

Patients were also given a leaflet regarding their low-back pain that included standardized advice and exercises.

A Chartered Physiotherapist trained in acupuncture carried out all treatments.


2) Placebo-TENS: Patients were advised that the treatment was relatively novel and that they should not

feel any discomfort with the procedure and, in fact, should not be aware of any sensation at all. They were

advised that the treatment had an effect on the nerve-endings and that it should relieve their symptoms.

Patient lying in the prone position for 30 minutes. A non-functioning TENS machine was attached to 4

electrodes placed over the lumbar spine and the unit was placed in a position to make it difficult to interfere

with the apparatus. The investigator monitored the patient’s condition after 10 and 20 minutes. Sessions: 6

over a 6-week period.

Patients were also given the advice and exercise leaflet and the same principal investigator carried out all

treatments.


Outcomes 1) Pain (VAS)

2) SF-36

3) Physical examination: finger-floor distance.

All these outcomes were measured immediately after the end of the 6th session.

4) Global improvement measured at 6 months: ”Did you experience pain relief? “Yes” or “No”. But only 40

(66.7%) patients were followed up to 6 months

Costs: not reported

Complications: In the acupuncture group there were 2/23 patients who reported side effects and 2/17 in the

placebo group.








06.01

06.02

06.04

06.07

Conclusions: “Although acupuncture showed highly significant differences in all the outcome measures

between pre and post-treatment, the differences between the two groups were not statistically significant”.


Study Kittang 2001

Methods - Randomization in blocks of four patients (method not described). No description of who performed the

allocation of patients

-Outcome assessor was blinded

-Funding: Three governmental, medical association and science council funding sources as well as funding

from two pharmaceutical companies

-Setting: Private clinic in Flora and Kinn, Norway

-Consent not described, ethics approval obtained

-57/60 patients were followed

-Analysis: t-test & Fishers exact test

Participants 60 patients with acute low-back pain (lasting less than 10 days).

Excluded: Neurologic outcomes, rheumatic illness, malign disease, systemic use of anti-inflammatory drugs

or steroids before inclusion and use of medicine that may interact with anti-inflammatory drugs.

Between 18 and 67 years of age

Gender: both sexes

Working status: 2/3 on sick leave at time of inclusion

Interventions 1) First treatment was needling in “lumbago 1 and 3” with medial lumbago, and in “upper lip” with more

lateral pain. Later treatments were 5 needles across at level L2, at “Ashi points” (local pain points) and in both

ankles. Analgesia was allowed and sick leave provided when necessary. Four treatments within two weeks.

Patients in both groups were given general advise and encouraged to daily physical activity.


2) Naproxen 500 mg twice daily for ten days


Outcomes 1) Pain (VAS) measured at baseline, 1 and 2 weeks and 3 and 6 months

2) Use of other analgesics measured at 1 and 2 weeks

3) Number of back pain episodes at 6 and 18 months

4) Side effects at 1 and 2 weeks

5) Stiffness measured at baseline, 1 and 2 weeks and 3 and 6 months

6) Lateral flexion measured at baseline, 1 and 2 weeks and 3 and 6 months

Costs: not reported

Notes Language: Norwegian


Asked authors for additional information: no response.





03.01

03.04

03.06

Conclusions: “No difference in reduction of pain or stiffness over a six-month evaluation”


Study Kurosu 1979(a)

Methods -Randomized (method not described). No information about concealment of allocation

-No information about blinding


-Setting: Private clinic in Tokyo, Japan

-There is no description about informed consent or ethics approval.

-Follow-up: 20 of 20 (100%)

-Analysis: Intention-to-treat, used t-test for between group analyses

Participants 20 patients with lumbar or sacral region pain.

Most of patients were between 40 and 50 years old.


Working status: Not reported

Previous treatments: Not reported

Interventions 1) Acupuncture: the needles were inserted, and left in situ for 10 minutes, and then removed. Insertion depth

was 2 to 4 cm, depending on one’s figure. Acupuncture needles used were stainless steel needles (50 mm

length, 0.25 mm diameter). Six to eight points in lumbar part were chosen from BL23, 24, 25, 26, 27, 31,

52 and 3 extra channel points by palpation. Abdominal needling was added: needles were inserted to a depth

of 1 to 1.5 cm at CV4,13 and ST25 (bilaterally). Acupuncture treatment was performed more than 4 times.


Experience: well-known and well-experienced acupuncturist.

2) Garlic moxibustion in lumbar region: Moxa is placed on top of a slice of garlic. Six to eight points in

lumbar area were chosen from BL23, 25, 27, 52 and the other points by palpation.


Outcomes 1) Pain: 10-item questionnaire about the specific actions that caused pain. Possible range of this questionnaire

is -10 to 20 (if patient feels pain at all actions) and higher scores are better. It was measured immediately

before second and fourth session

Costs: not reported




No additional information from authors.


11.02

Conclusions: “There is no difference between needle retention technique and garlic moxibustion for low-

back pain”.





Study Kurosu 1979(b)

Methods -Randomized (method not described). No information about concealment of allocation

-No information about blinding


-Setting: Private clinic in Tokyo, Japan

-There is no description about informed consent or ethics approval.

-Follow-up: 20 out of 20 (100%)

-Analysis: Intention-to-treat, used t-test for between group analyses

Participants 20 patients with lumbar or sacral region pain.

Most of patients were between 40 and 50 years old.


Working status: Not reported

Previous treatments: Not reported

Interventions 1) Acupuncture: the needles were left in situ for 10 minutes, and then removed. Depth was 2 to 4 cm,

depending on one’s figure. Stainless steel needles (50 mm length, 0.25 mm diameter). Six to eight points

in lumbar part were chosen from BL23, 24, 25, 26, 27, 31, 52 and 3 extra channel points by palpation;

abdominal needling was added: needles were inserted to a depth of 1 to 1.5 cm at CV4,12 and ST25

(bilaterally). Acupuncture treatment was performed more than 4 times.

Experience: well-known and well-experienced acupuncturist.


2) Other acupuncture technique: needles were removed immediately after insertion. Insertion depth was

2 to 4 cm, depending on one’s figure. Stainless steel needles (50 mm length, 0.25 mm diameter). Six to

eight points in lumbar part were chosen from BL23, 24, 25, 26, 27, 31, 52 and 3 extra channel points by

palpation. Abdominal needling was added: needles were inserted to a depth of 1 to 1.5 cm at CV4,12 and

ST25 (bilaterally); needles were left in situ for 10 minutes, and then removed. Acupuncture treatment was

performed 3 times.


Outcomes 1) Pain: 10-item questionnaire about the specific actions that caused pain. Possible range of this questionnaire

is -10 to 20 (if patient feels pain at all actions) and higher scores are better. It was measured immediately

after the fourth session

Costs: not reported




No additional information from authors.


12.02

Conclusions: “Results of needle retention technique is superior to that of simple insertion technique for low-

back pain”.





Study Lehmann 1986

Methods -Block randomization, blocks defined by prior lumbar surgery (method not reported). No information about

concealment of allocation.

-Therapists were blinded between real TENS and sham TENS, but not between acupuncture and TENS

-Funding: NIHR Grant

-Setting: Multidisciplinary inpatient clinic in a University of Iowa Hospital, USA.

-Informed consent and ethics approval were not reported

-Follow-up: 39 of 54 randomized patients (72%)

-Analysis: Multivariate analysis of covariance (adjustments for baseline scores and for non-organic signs). No

intention-to-treat analysis.

Participants 54 patients screened at orthopaedic clinic with chronic (>3 months) disabling low-back pain.

Excluded: candidates for lumbar surgery, pain less than 3 months, pregnancy, osteomyelitis of the spine,

discitis, tumour, ankylosing spondylitis, vertebral fractures and structural scoliosis.

Diagnoses: chronic disabling (not working) low-back pain. Duration of low-back pain: 48% more than 18

months.

Mean age: 39 years old (ranged from 20 to 59)

Gender: 33% females.

93% married.

Working status: 1/54 was working. 51 were receiving compensation. 33% were involved with litigation.

Previous treatments: some had surgery.

Interventions 1) Electroacupuncture with needles, biphasic wave at 2 to 4 Hz, inner and outer bladder meridian for

paravertebral pain. Gall bladder meridian for lateral (sciatic) pain. LI4 points and additional points were

stimulated according to the patient’s pattern of pain; certified and experienced acupuncturist; twice weekly

for 3 weeks. Teh Chi not reported.


2) Real TENS, pulse width of 250/second at 60 Hz, 15 treatments in 3 weeks, sub-threshold intensity, points

of stimulation over the center of pain, experienced physiotherapist.


3) Sham TENS, same as TENS but dead battery.


Outcomes 1) Peak pain and average pain (VAS)

2) Activities of daily living: 15 items (yes/no)

3) Physician’s perception of improvement

4) Range of motion

All these outcomes were measured at baseline, at discharge and between 3 to 6 months after discharge

5) Return to Work after 6 months (from no disability=10 points, to not able to work at all=0 points);

Costs: not reported

Complications: there were no complications.





06.06

06.08





07.07

07.08


Conclusions: “There were no significant differences between treatment groups with respect to their overall

rehabilitation”. “The electroacupuncture group demonstrated slightly better results than the other groups.”


Study Leibing 2002

Methods -Computer-based randomization method.

-Patients and outcome assessors were blinded. (Patients were blinded only between two types of acupuncture)

-Funding: Ministry of Education, Science, Research and Technology, Federal Republic of Germany.

-Setting: Outpatient clinic. Department of Orthopaedics, University Goettingen, Germany

-Informed consent and ethics approval were obtained.

-Follow-up: 150 patients were randomized. 131 initiated treatment. 114 (76%) were followed at the end of

the treatment and 94 (63%) at 9 months.

-Analysis: ANOVA with post-hoc comparisons using Tukey studentized range tests when significant overall

effects observed. No intention-to-treat analysis, but used last observation carried forward from the 131

patients that initiated treatment.

Participants 150 patients with chronic (> 6 months) non-radiating low-back pain.

Excluded: Abnormal neurological status, concomitant severe disease, psychiatric illness, current psychother-

apy, pathological lumbosacral anterior-posterior and lateral X-rays (except for minor degenerative changes),

rheumatic inflammatory disease, planned hospitalization and refusal of participation.


Gender: 58% female

76% married

Mean BMI: 26.3

Working status: 82% employed

Current treatments: 8.4% surgery. 50% analgesics

Interventions 1) All patients received standardized active physiotherapy of 26 sessions (each 30 minutes) over 12 weeks. It

was performed by trained physiotherapists according to the Bruegger-concept. In addition, 20 sessions (each

30 minutes) by an experienced Taiwanese physician over 12 weeks. In the first 2 weeks, acupuncture was

done 5/week, and in the next 10 weeks, 1/week.

Combined traditional body and ear acupuncture. Twenty fixed body acupoints (9 bilateral, two single points)

and six on the ear (alternately on one ear) were selected according to their function in TCM and were needled

in every patient. No diagnostic procedure was done to determine individual acupoints.

Body points were manually stimulated until Teh Chi and left in place for 30 minutes: GV3, GV4, BL23,

BL25, BL31, BL32, BL40, BL60, GB34, SP6, Yautungdien (extra meridian, at the back of the hand).

Ear points (left in for one week): 38, 51, 52, 54, 55, 95

Randomized to this group: 50, but only 40 initiated treatment. Ten were lost before first session. Reasons:

withdrew consent=3; exclusion criteria appeared prior to treatment=5; relocated=2.

2) No additional treatment. Only active physiotherapy (as described above)

Randomized to this group: 50, but only 46 started treatment. Four were lost before first treatment. Reasons:

withdrew consent=2; exclusion criteria=2.

3) Sham acupuncture plus physiotherapy. Sham acupuncture received 20 sessions (each 30 minutes) of

minimal acupuncture by the same physician over 12 weeks. Sham acupuncture was done following the




standards of minimal acupuncture. Needles were inserted superficially, 10 to 20 mm distant to the verum-

acupoints, outside the meridians, and were not stimulated (no Teh Chi).

Randomized to this group: 50, but only 45 started treatment. Reasons: withdrew consent=1; exclusion

criteria=4.

Outcomes 1) Pain intensity: 10 cm VAS

2) Pain disability: total score consists of 7 areas of activity (min 0, max 70) O=no disability, and 70=total

disability.

3) Psychological distress: Hospital Anxiety and Depression Scale, 14-item instrument for use in non-psychi-

atric medical patients. Total score (0 to 42) is a measure of psychological distress.

4) Spine flexion, fingertip-to-floor distance (min = 0 cm)

Costs: not reported

Complications: minor, not serious adverse events occurred in three patients in the acupuncture group.

Notes The use of last observation carried forward usually attenuates the differences between groups.

Language: English




06.01 06.03

06.08 06.12

06.13 06.15

13.01 13.02

13.03 13.05

13.08 13.09

Conclusions: “Acupuncture plus physiotherapy was superior to physiotherapy alone regarding pain inten-

sity, disability and psychological distress at the end of the treatment. Compared to sham acupuncture plus

physiotherapy, acupuncture (plus physiotherapy) reduced psychological distress only. At 9 months, the su-

periority of acupuncture plus physiotherapy compared to physiotherapy alone became less and acupuncture

plus physiotherapy was not different from sham plus physiotherapy”.


Study Li & Shang 1997

Methods -Randomized (method not reported).

No mention of concealment of allocation.

-Patients were blinded. Comment: since both groups were given active treatments, all the patients should know

that they were treated by “real” acupuncture. However, they probably couldn’t tell which active treatment

group they were in.


-Setting: Outpatient clinic in a hospital. Hebei Province, China.

-Informed consent and ethics approval were not mentioned

-Follow-up: all 156 patients were followed.

-Analysis: U-test: between groups

Participants 156 patients with low-back pain of varying duration (between 2 days and 8 years)

Diagnoses: not reported

Excluded: not reported

Age between 20 and 71 years old







Interventions 1) Manual acupuncture plus cupping. Teh Chi sensation was obtained and needles were retained for 20

minutes. Major points: BL23, 40. GV 2, 26, LU5. Supplement points: for coldness and dampness: GV3,

BL31, 34. For blood stasis: BL17, 18. For kidney deficiency: GV4 and KI 3. Treatment was given every other

day (except for acute back pain, which was treated daily) up to 10 treatments.


Experience: adequate

2) Manual acupuncture alone. Major points: BL23, 40 and GV2. Supplement points: same as treatment

group.


Outcomes 1) Overall assessment (see description in He 1997). Measured immediately after the end of the sessions.

Costs: not reported


Notes Language: Chinese





Conclusions: “Manual acupuncture plus cupping technique is better than manual acupuncture alone for

treating low-back pain”


Study Lopacz & Gralewski

Methods - Randomization procedure not described.

- Nobody was blinded.

Participants 34 male patients from a neurology department.

Inclusion criteria: low-back pain for 1 month or more.

Age: mean 42 years old (ranged from 25 to 52).

Interventions 1) Acupuncture: 4 needles close to spine, 10 minutes, 4 treatments, 8 days, plus pharmacotherapy. Teh Chi

unclear. Training & experience of acupuncturists unknown.


2) Placebo, suggestion, new Swedish method for pain relief, same 4 points echo-encephalography, 10 minutes,

4 treatments, 8 days, plus pharmacotherapy.


Outcomes 1) Global improvement (5-point scale): very good, good, doubtful, unchanged and worsening.

Measured after first treatment and after 4 treatments

Costs: not reported


Notes Very short term follow-up only. Small sample size.

Authors dichotomized at very good + good versus others.

We classified the patients as chronic low-back pain.




Language: Polish




06.02

Conclusions: “The therapeutic results were better, both immediately and after a series of acupuncture. The

difference in the results of treatment was statistically significant in the patients with longest duration of pains

(>3 months)”.


Study MacDonald 1983

Methods -A stratified random process to divide the sexes as equally as possible between the two groups.

-Patients and observers were blinded.

-Funding: North West Thames Regional Health Authority

-Setting: London

-Informed consent and ethics approval not reported

-Follow-up: not reported

-Analysis: Wilcoxon rank sum test.

Participants 17 patients referred from orthopaedic or rheumatological departments.

Inclusion criteria: chronic LBP for at least one year, no relief from conventional treatments.

Diagnoses: spondylitis, ankylosing spondylitis, degenerative disc lesion, idiopathic, non-articular rheuma-

tism, osteoarthritis, prolapsed intervertebral disc, arachnoiditis, ligamentous strain and Scheuermann’s os-

teochondritis.

Exclusion criteria: not reported

Demographics: not reported. But it says ”the two groups were comparable in terms of age, duration of pain,

mood scores, number of physical signs and severity of pain.

Interventions 1) Superficial needling: subcutaneous (4 mm) 30-gauge needle insertion at trigger points. (Number of trigger

points unknown). 5 to 20 minutes, maximum of 10 treatments in 10 weeks. Electrical impulses 700µs at 2

Hz if manual stimulation failed. Randomized to this group: 8

Experience: unknown

2) Placebo transcutaneous electrical stimulation: electrodes connected to dummy apparatus, maximum 10

treatments in 10 weeks.


Outcomes 1) Pain relief:

- worse (-1)

- no change (0)

- minimal improvement (1% to 24%) (1)

- moderate improvement (25% to 49%) (2)

- good (50% to 74%) (3)

- excellent (75% to 99%) (4)

- complete resolution (100%) (5)

2) Pain score reduction

3) Activity pain score reduction

4) Physical signs reduction

5) Severity and pain area reduction

Costs: not reported





Notes Intervention is dry-needling.

Very small sample size, number of treatments unknown, and follow-up time unknown.

Language: English



Results:

1) Pain relief: dry-needling: 77.36, placebo: 30.14 (p<0.01);

2) Pain score: dry-needling: 57.15, placebo 22.71 (p:NS);

3) Activity: dry-needling 52.04, placebo 5.83 (p<0.05);

4) Physical signs: dry-needling: 96.78, placebo: 29.17 (p<0.01);

5) Severity and pain area: dry-needling: 73.75, placebo: 18.89 (p<0.01);

Conclusions: “Needling achieved better responses than the placebo in all five measures. Four of the five inter-

group differences were statistically significant.”


Study Mendelson 1983

Methods -Randomized (method not described). Unclear about concealment of allocation. Cross-over study.


-Funding: National Health and Medical research Council of Australia.

-Setting: Prince Henry’s and Alfred Hospitals, Melbourne, Australia.

-Informed written consent was obtained. Ethics approval by the Ethics Review Committee.

-Follow-up: 77 of the 95 patients randomized (81%).

-Analysis: T-tests. No intention to treat analysis.

Participants 95 volunteers with chronic low-back pain, no compensation or litigation pending, no overt psychiatric

disease.

Diagnoses: Osteoarthritis, traumatic spondylopathy, disc lesion, sacroiliac joint disorder and backache not

specified.


Gender: 37 males and 40 females.

Pain duration: 12 years.

Interventions 1) Traditional Chinese acupuncture by a surgeon trained in Peking; points: B23, 25, 36, 40 and 60. If

sciatica: GB 30, 34 and 39. Average 8 needles, manual stimulation until reaching Teh Chi, 30 minutes with

no further stimulation, twice weekly, 4 weeks.

Randomized to this group: don’t know. 36 completed the study.

2) Sham acupuncture, intradermal injection of 2% lidocaine at non-acupuncture, non-tender sites, then

acupuncture needles superficially into the infiltrated areas for 30 minutes without stimulation, twice weekly,

4 weeks.

Randomized to this group: don’t know. 41 completed the study.

Outcomes 1) Pain (VAS) 100-mm scale.

2) Pain relief

3) McGill Pain Questionnaire

4) Disability (method not described)

Costs: not reported









06.01

06.02

Conclusions: “Patients receiving acupuncture had a greater but not significantly different reduction in pain

rating scores compared with those receiving placebo. Similarly, no significant difference was found between

the two groups based on self-assessment of disability”.


Study Meng 2003

Methods -Randomized (computer generated random allocation sequence). Serially numbered, sealed, opaque envelopes

-No blinding

-Funding: New York Chapter of the Arthritis Foundation

-Setting: Private surgeries clinics of the Hospital for Special Surgery at the New York Presbyterian Hospital.

USA

-Written informed consent; approval by Institutional Review Board

-Follow-up: 47 of 55 randomized patients (85%)

-Analysis: ANOVA for between group differences. Both intention-to-treat and completers only analysis

Participants 55 patients with chronic nonspecific low back pain (>12 weeks) and older than 60 years.

Excluded: specific cause for low-back pain, prior use of acupuncture, use of corticosteroids, muscle relaxants,

narcotics, anticoagulants.


Gender: 22 male and 33 female

Previous treatments: 27 NSAID, 10 analgesics, 1 muscle relaxant and 2 aspirin.

Ethnicity: 47 Caucasian, 5 African-American and 3 Hispanic.

Charlson Comorbidity Index: 1.7 (+/- 2.0)

Interventions 1) Acupuncture plus standard therapy: Acupuncture twice a week for 5 weeks. Total 10 sessions. 30-gauge

needles with electrical stimulation (4 to 6 Hz) with a pulse duration of 0.5 ms. Teh Chi response at all

points were verified. Between 10 and 14 needles were used per session. Needle retention was 20 minutes.

Fixed acupoints: UB23, 24, 25, 28 (bilateral). Du3 and 4. Supplementary acupoints: maximum 4 additional

needles: UB36, 54, 37, 40, GB 30, 31.

Two anaesthetists certified in acupuncture.

Randomized to this group: 31. Received acupuncture: 28. Completed follow-up: 24.

2) Standard therapy: Primary physician for 5-week intervention period: NSAID, aspirin, non-narcotic anal-

gesic. Continue back exercise (physical therapy) or home exercise regimen. Prohibited: narcotics, muscle

relaxants, TENS, epidural steroid injections and trigger point injections.

Randomized to this group: 24. Received standard therapy: 23. Completed follow-up: 23.

Outcomes 1) Back specific functional status (modified Roland Disability Questionnaire)

2) Pain (VAS)

These outcomes were measured at 0, 2, 6 and 9 weeks during the trial period, but we only used the measures

at 6 weeks (at the end of all sessions) and 9 weeks (3 to 4 weeks after the end of the sessions) .




Costs: not reported

Complications: no difference in adverse effects.




13.01

13.02

13.05

13.07

Conclusions: “Our data indicate that acupuncture plus standard therapy does decrease back pain and disability

in older patients compared with standard therapy alone in a clinically and statistically significant manner”


Study Molsberger 2002

Methods -Randomized (computer generated) stratified according to pain duration. Central telephone randomization.


-Funding: Grant from the German Ministry of Education, Science and Research

-Setting: Inpatients in the Hospital. Dusseldorf, Germany.

-All patients were informed about the trial and written consent was obtained.

-Follow-up: 124 of 186 patients randomized (66%)

-Analysis: Approximate chi-square or exact Fisher test, non-parametric Mann-Whitney-Wilcoxon rank test.

Per protocol analysis (n=174) and intention-to-treat analysis (n=186). Main analysis is adjusted for multiple

testing.

Participants 186 patients with low-back pain lasting longer than 6 weeks, with average pain scores greater than 50 mm

(max 100 mm) during the last week. Aged between 20 and 60 years old, and speak German.

Excluded: sciatica, neurological disorder, disc or spine surgery, bone or joint disorder, previous treatment

with acupuncture, psychiatric illness, pregnancy, regular intake of analgesics, off work longer than 6 months,

no litigation.



Mean duration of pain: 9.9 years.



Interventions 1) Verum acupuncture plus conventional orthopedic therapy. Acupuncture: standard points: BL23, 25, 40

and 60 and GB30 and 34. In addition, up to four points of maximum pain “Ah shi points”, which were often

close but not necessarily identical to BL 54, 31, 32 were needled. Needle insertion ranged from 1 to 10 cm

and needle manipulation was mild to strong. A Teh Chi feeling was always achieved. During the acupuncture

treatment, no additional treatment was administered. All patients received 12 verum acupuncture treatments,

3/week, each lasting for 30 minutes.

Acupuncture was carried out by an experienced medical doctor, who had studied in China (Beijing).

Randomized to this group: 65. Drop-outs during treatment: 7. Lost to follow-up: 11.

2) Sham acupuncture plus conventional orthopedic therapy. Sham acupuncture received 12 sham acupuncture

treatments, 3/week, each lasting 30 minutes. Sham acupuncture was standardized to ten needles applied

superficially (depth of needle insertion was less than 1 cm) at defined non-acupuncture points of the lumbar

region, and five needles on either side of the back.




Randomized to this group: 61. Drop-outs during treatment: 3. Lost to follow-up: 17

3) The conventional orthopedic therapy consisted of: daily physiotherapy, physical exercises, back school,

mud packs, infrared heat therapy. On demand they received 50 mg diclofenac up to three times a day.

Injections or cortisone application of any kind were not allowed.


Outcomes 1) Pain intensity (VAS) during the last 7 days.

2) At least 50% reduction in pain intensity

3) Effectiveness of treatment: excellent, good, satisfactory and failed. Dichotomized at exc+good versus

satisfactory+failed.

4) Schober and finger-to-floor distance.

All outcomes were taken at the end of the treatment period and 3 months later.

Costs: not reported

Complications: no side effects or complications occurred in any treatment group





06.01

06.02

06.14

13.01

13.04

13.06

Conclusions: “Together with conservative orthopedic standard therapy, acupuncture helps to decrease pain

intensity directly after treatment, and patients’ rating of the acupuncture treatment is significantly better

than that of the standard therapy alone. The therapeutic effect lasts for at least 3 months after the end of

treatment”


Study Sakai 1998

Methods -Randomized (method not described). Sealed envelopes.

-Not blinded.

-Funding: Grant from the Foundation for Training and Licensure Examination in Anma-Massage-Acupres-

sure, Acupuncture and Moxibustion.

-Setting: Outpatients in a University Hospital. Tokyo, Japan.

-Informed consent was taken orally. No description of ethics approval.

-Follow-up: no information

-Analysis: No statistical test.

Participants 26 patients with non-specific low-back pain of variable duration

Excluded: (1) osteoarthritis of lumbar spine, osteoporosis, scoliosis, spondylolysis, spondylolisthesis, herni-

ation of lumbar disc, spinal stenosis or fracture; (2) radiculopathy or neuropathy in the lower extremity;

(3) urological problem, gynaecological problem, neurological problem, collagen, diabetes or malignancy; (4)

increase of CRP or ESR; (5) medication of corticosteroid, immunosuppressant agent, NSAID or muscle

relaxant; (6) problem of general condition; (7) dementia; (8) pregnancy; (9) elderly patient; (10) those who

were judged to be inappropriate for the trial by the authors.





Gender: 7 males and 19 females



Interventions 1) Needling points in lumbar area were chosen from BL23, 25, 32, 52 and 2 extra channel points near the

spinous process of L4 and L5, and that in lower extremity were chosen from BL37, 40, 57, ST36, GB34

by palpation. Manual acupuncture technique such as needle retention and sparrow pecking technique were

performed. Electro-acupuncture was applied in some cases. Other details in technique not reported. Patients

were treated twice a week for two weeks, i.e. four sessions in total.


Experience: unknown

2) Oral medication, which includes NSAID and/or kampo medicine (Chinese herbs).


Outcomes 1) Pain relief (VAS) on average on the pervious day, rated by the patient. Higher values are better.

2) JOA (Japan Orthopaedic Association) Score rated by the physician. Subjective symptoms of back pain

and restriction of daily activities. Maximum 17 points. Higher values are better.

Outcomes were measured Immediately at the end of all sessions.

Costs: not reported


Notes This study reports on two distinct trials, but we used only the data from one trial, which was randomized.

The other trial allocated patients using alternation, therefore it was not randomized.

Language: Japanese

Publication: abstract (and unpublished report).

We obtained additional information from authors


11.01

11.04

Conclusions: “Results of acupuncture are the same as those of medication for low-back pain”


Study Sakai 2001

Methods -Multicentric study.

-An independent controller in central office prepared an allocation table and sealed envelopes. A computer-

generated randomized allocation table was used to make a sequence of sealed opaque envelopes containing

the code of intervention. The assigned envelope was opened by acupuncturist at the patient’s entry into the

trial

-Outcome assessor was blinded.


sure, Acupuncture and Moxibustion.

-Setting: Outpatients in a University Hospital. Tokyo, Osaka, Kyoto and Tsukuba, Japan.

-Written informed consent was taken from patients. At two of the four centers, judgement of ethics committee

was asked and the committees approved the protocol. The other two centers did not have ethics committees.

-Follow-up: 64 of 68 randomized (94%)

-Analysis: 95%CI and repeated measures ANOVA.

Participants 68 patients with low-back pain (at least 2 weeks) and age 20 years or older.




Diagnoses: lumbago (22), lumbar spondylosis (15), discopathy (9), acute low-back pain (3), spondylolysis

(3) , spondylolisthesis (1), sacroiliitis (1) and unclassified (10).



problem or cardio-vascular problem; (6) patients who can not stop other conflicting or ongoing treatments;

(7) problem of general condition; (8) dementia; (9) pregnancy; (10) other patients who were judged to be

inappropriate for participating in the trial.


Gender: 35 females and 29 males.



Interventions 1) Needling points were chosen by palpation from the part of quadratus lumborum (around BL52) and/or

erector spinae (around BL23 and BL26) in the lumbar area. Two points were used bilaterally - in total four

points - for each treatment. Patients were treated twice a week for two weeks.

Two types of disposable stainless steel needles were used according to patient’s stature and fat: 0.20 mm in

diameter and 50 mm in length, and 0.24 mm in diameter and 60 mm in length. Needles were inserted into

the muscles. Electro-stimulation at frequency of 1 Hz was applied for 15 minutes. The intensity was adjusted

to make muscle contraction without pain.


Experience: unknown

2) TENS: Same points as above. Two points were used bilaterally - in total four points - for each treatment.

Patients were treated twice a week for two weeks, i.e. four sessions in total.

Gel type disposable electrodes of 20 x 30 mm in size were used. Stimulation with the frequency of 1 Hz was

applied for 15 minutes


Outcomes 1) JOA (Japan Orthopaedic Association) Score rated by the physician. Subjective symptoms of back pain

and restriction of daily activities. Maximum 20 points. Higher values are better.

2) Pain relief (VAS) on average on the pervious day, rated by the patient. Higher values are better.

These outcomes were taken after the end of the 4 sessions.

Costs: not reported

Complications: no adverse event was reported in the electroacupuncture group. In the TENS group: 1 itching

and 1 dullness after session.

Notes Duration of low-back pain mixed.

Language: English and Japanese


We obtained additional information from the authors.


11.01

11.03

11.04

11.07

Conclusions: “There was no significant difference between groups in any parameter”





Study Takeda & Nabeta 2001

Methods -Randomized (using draws). Stratified by pain duration and gender. Using sealed and numbered envelopes,

but the person doing the randomization was not independent.

-Patients blinded.

-Funding: no funding was received.

-Setting: Acupuncture College in Osaka, Japan.

-Informed consent was obtained from participants and there was no description of ethics approval.


-Analysis: Mann-Whitney U test for between group differences. No intention-to-treat analysis.

Participants 20 students of acupuncture college who were suffering from lumbago.

Excluded: sciatica

Duration of pain: Mean 40.4 months in distal group and 81.0 months in local group.

Mean age: 26.4 years old in distal group and 35.8 years in local group.


Working status: all students.


Interventions 1) Distal point technique: At the acupuncture points in lumber area: BL23, 26 and Yao-yan (extra-point: EX-

B7), acupuncturist mimicked needle insertion: tapped head of needle guide tube, then gesture of needling

was performed. Acupuncture points in lower extremity: BL37, 40 and 58, were needled by real acupuncture

needle (40 mm in length and 0.2 mm in diameter). Insertion depth was 1 to 2 cm. Sparrow-picking technique

was performed 5 times, then needles were removed. Participants were treated once a week for 3 weeks.

Experience: unknown

Randomized to this group: 10. Drop-outs during study: 1.

2) Local points technique: Acupuncture points in lumber area: BL23, 26 and Yo-gan (extra-point: EX-B7),

were needled by real acupuncture needle (40 mm in length and 0.2 mm in diameter). Insertion depth was 1

to 2 cm. Sparrow-picking technique was performed 5 times, then needles were removed. At the acupoints in

lower extremity: BL37, 40 and 58, acupuncturist mimicked needle insertion: tapped head of needle guide

tube, then gesture of needling was performed. Participants were treated once a week for 3 weeks.

Experience: unknown

Randomized to this group: 10. Drop-out during treatment: 1.


2) Function: activity of daily living score. 8 questions about difficulty of specific actions. Maximum 16 points.

Higher values are better.

3) Finger-to-floor distance.

All these outcomes were measured immediately before and after the treatment.

Costs: not reported



Publication: abstract only



12.01

12.04

12.05




Conclusions: “There is no difference between the effects of lumbar area needling and that of distal point

needling”


Study Thomas & Lundberg 94

Methods - Randomized (method not described). No description of concealment of allocation.

- Outcome assessors were independent and not involved in the treatment.

- Funding: Karolinksa Institute Foundation, King Gustav Vth 80 year anniversary Fund, Tore Nilssons

Foundation for Medical Research, Torsten and Ragnar Soderbergs Foundation and The Swedish Medical

Research Council.

- Setting: Outpatient clinic at the Karolinska Hospital. Stockholm, Sweden.

- Oral informed consent was obtained. No description of ethics approval:

- Follow-up: 40 of 43 randomized patients (93%)

- Analysis: Student t-test for independent samples and multiple comparisons ANOVA. No intention-to-treat

analysis.

Participants 43 patients from 2 clinics with nociceptive LBP for 6 months or more, restriction of trunk or hip movement

due to pain, restriction of ADL, muscle spasm.

Excluded: previous surgery, claudication, depression, neurosis, clinical examination not nociceptive.

Diagnoses: Osteoarthritis, sacroiliac joint, sciatica, intervertebral disc degeneration, disc prolapse, lumbar

strain, osteoporosis

Demographics and patients characteristics: not reported, but they say there were no significant differences

between the groups.

Interventions 1) Acupuncture: three different modes of acupuncture: a) manual stimulation, b) low frequency (2 Hz) and

c) high frequency (80 Hz) electrical stimulation of needles. Six local points (3 pairs of paraspinal points: UB

23, 25, 26 or 32) and 3 to 4 distal points (SI 6, UB40 or 60, GB 30 or 34 or St36). Insertion 1 to 5 cm,

rotation producing Teh Chi, 10 sessions of 30 minutes; 2 registered physiotherapists trained in acupuncture.


2) Waiting list controls, no treatment.


Outcomes 1) Pain: number of words from chart of 83 words describing pain intensity

2) Global improvement: 3-point scale (improved, no change, worse)

3) Functional status: VAS on 12 ADL . Results are presented as number of activities that cause less than 50%

pain.

4) Mobility: goniometry of the lumbar spine

Outcomes were measured after 6 weeks and 6 months.

Costs: not reported


Notes Randomization only for comparison acupuncture versus WLC, not for different modes of acupuncture.

Language: English



Results see comparisons:

05.01

05.03

05.07




The results for global improvement could not be entered in the graphs.

The authors found significant differences for pain outcomes, however, when we entered this data in RevMan

(comparison 05.01) this was not significant. Because we did this based on the data extracted from the figure,

we decided to follow the author’s conclusions.

Functional outcomes had to be transformed to effect sizes (comparison 05.07 to be combined with another

trial)

Conclusions: “After 6 weeks, patients receiving acupuncture were statistically significantly better than the

control group on measures of pain, global improvement and mobility. The same results were observed at 6

months, but only for the group that received low frequency electroacupuncture”.


Study Tsukayama 2002

Methods -Randomized. Computer-generated random number were used to make a sequence of sealed envelopes. An

independent person prepared an allocation table and sealed envelopes.



sure, Acupuncture and Moxibustion” and the Tsukuba College of Technology.

-Setting: Private clinic in Tsukuba, Japan.

-The study was approved by the Ethics Committee of TCT Clinic. Informed consent was taken from patients

according to the ICH/GCP.

-Follow-up: 19 of 20 patients (95%)

-Analysis: Repeated measures ANOVA. No intention-to-treat analysis.

Participants 20 patients with low-back pain of at least 2 weeks and over 20 years old.

Excluded: radiculopathy or neuropathy, fracture, tumour, infection or internal disease, other general heath

problem and conflicting or ongoing treatments.

Duration of low back pain: acupuncture group=2900 days (+/- 1983) and TENS group=3120 days (+/-

3306).



Working status: not described

Previous treatments: acupuncture (4)

Interventions 1) Acupuncture: Points selected by tenderness and palpable muscle bands detected on the lower back and

the buttock. Four points bilaterally (8 in total) were used for each treatment. Points most frequently used

were BL23 and BL26. Two types of disposable stainless steel needles were used, depending on stature and

fat: 0.20 mm in diameter and 50 mm in length and 0.24 mm in diameter and 60 mm in length. Needles

were inserted into the muscles. The average insertion depth was approximately 20 mm. Electrostimulation

was applied to the inserted needles with an electronic stimulator with a frequency of 1 Hz for 15 minutes.

Press tack needles were inserted after EA at four of the 8 chosen points and left in situ for several days, they

are 1.3 mm long projecting from the sticky side of a small round adhesive dressing.

Patients were treated twice a week for 2 weeks, for 4 sessions in total.

Randomized to this group: 10. Drop-outs: 1

Experience: unknown

2) TENS: Gel type disposable electrodes of 20x30 mm were used for 8 points. Electro-stimulation was

applied in the same manner as in the acupuncture group. The intensity was adjusted to the maximum

comfortable level, and muscle contraction was observed. After each session, a poultice containing methyl

salicylic acid, menthol and antihistamine was prescribed to be applied to the low-back region, at home, in-

between treatments.




Patients were treated twice a week for two weeks, in total 4 sessions.

Randomized to this group: 10. No drop-outs.

Outcomes 1) Pain (VAS): average pain level on the previous day.

2) JOA score. See description in Sakai 2001

These outcomes were measured 3 days after the last session.

Costs: not reported

Complications: no adverse events reported by the evaluator. The therapists reported transient aggravation of

symptoms in the acupuncture group (1), discomfort due to tack needles (1), pain on needle insertion (1)

and small subcutaneous bleeding (1). In the TENS group: transient aggravation (1), transient fatigue (1) and

itching (1).





07.11

07.08

Conclusions: “The results of the present trial showed a significant between-group difference in pain relief in

favour of acupuncture”


Study Von Mencke 1988

Methods -Randomization procedure not described.

-Patient and outcome assessors blinded.

-Setting: Secondary care.

Participants 65 patients from an orthopedic clinic with lumbago and/or ischias, no relief after conventional treatment.

Diagnoses: Lombociatalgia (30), low-back pain (20), LWS Syndrome (10) and Ischialgia (5).

Exclusion criteria: neurological problems, scoliosis, concurrent treatment, acute disc prolapse or protrusion,

chronic degenerative disorders, infection.

Age and gender: not described.

Heterogeneous population regarding type, location and duration of disorder.

Interventions 1) Manual acupuncture, traditional meridian acupuncture or trigger points, rotation, insertion 0.2 to 3 cm,

6 to 12 needles 5 to 20 minutes, 8 treatments. Training & experience of acupuncturists unknown.

Points:

- Posterior: GV20, BL26, 31,33, 35, 48, 50, 54, 57, 58, 60.

- Lateral: GV20, GB 26, 28, 30, 32, 34, 37, 38, 40. BL 26, 31, 33, 48, 60

- Anterior: GV 20, ST 36, 40. BL 31, 33, 48, 60.


2) Sham acupuncture, no traditional acupuncture nor trigger points.



2) Global improvement

3) Schober’s test

4) Lasegue’s test

Notes Language: German






Results:

1) Improvement in pain at short-term follow-up: acupuncture=55%; sham acupuncture=37%. Long-term:

44% versus 30%.

2) Global improvement: acupuncture=94%, sham acupuncture=50% (Table 10.02).

3) Increase in Shober test: short-term: acupuncture=6.4, sham acupuncture=2.7. Long-term: 7.8 versus -0.9

4) Lasegue: short-term: acupuncture=6.0, sham acupuncture=2.2. Long-term, acupuncture=6.7, sham

acupuncture=0.6

Conclusions: “The difference in improvement between typically and atypically treated patients was highly

significant (p<0.0001).”


Study Wang 1996

Methods -Randomized (method not reported). No description of concealment of allocation.

-Patients were blinded.


-Setting: Not reported. Vanuatu, Southwest Pacific Ocean.

-Informed consent and ethics approval: Not mentioned

-Follow-up: not described but it seems 100%.

-Analysis: U-test. No intention-to-treat analysis

Participants 492 patients with low-back pain of unknown duration.

Diagnoses: back pain

Exclusion criteria not reported.

Mean age: 48% were older than 40 years old.




Interventions 1) Local treatment plus cupping. Teh Chi sensation was obtained and needles were retained for 20 minutes.

Points: BL23, 25 and 32. Treatments were given daily up to 10 treatments.


Experience: unknown

2) Distal treatment plus electrical stimulation. Points: ST36, GB 39, BL60 and LI4.


Experience: unknown

Outcomes 1) Overall assessment: a) cure: no pain and normal range of motion, no tenderness upon palpation, and

normal life and work status. b) effective: pain is markedly improved, normal lumbar movement, no obvious

tenderness upon palpation, and life and work is not affected c) no significant change.

Measured 3 months after the sessions.

Costs: not reported


Notes The authors dichotomized at : Cure+effective versus no change.

Language: Chinese







12.03

Conclusions: “Local acupuncture treatment plus cupping is more effective (p<0.05) than the distal treatment

plus electrical stimulation.”


Study Wu (a) 1991

Methods -Randomized (based on odd or even number of the date of patient admission). No mention of concealment

of allocation

-Patients were blinded.


-Setting: Outpatients in a hospital. Morocco.

-Informed consent and ethics approval not mentioned.

-Follow-up: 100% (single session of acupuncture)

-Analysis: Not reported

Participants 150 patients with acute low-back pain.

Exclusion criteria not described

Age between 20 and 55 years old




Interventions 1) SI3 point treatment


2) Extra 29 (EX-UE7) treatment


Manual acupuncture technique (no electro-stimulation) was used. Strong Teh Chi sensation was obtained

combined with lumbar spine movement until symptom relieved. No mention of the duration of the treatment.

Outcomes 1) Global assessment (pain and range of motion).

- cure: no pain and normal range of motion

- marked effective: pain is generally gone and ROM marked improved

- effective: pain is relieved and ROM is somewhat improved.

- no change

Costs: not reported


Notes The authors dichotomized at:

a) cure+marked effective+effective versus no change

and

b) cure+marked effective versus effective+no change

Language: Chinese







Dichotomization a) 04.02

Dichotomization b) 04.02

Conclusions: “Acupuncture point SI 3 is more effective than the point Yaotongxue.”


Study Wu (b) 1991

Methods See Wu (a) 1991

Participants

Interventions

Outcomes

Notes


Study Yeung 2003

Methods -Randomized in blocks (method not described). Randomization was blinded.

-Outcome assessors blinded.

-Funding: The Hong Kong Development Fund and Tung Wah Board Fund

-Setting: Outpatient clinic in a hospital. Hong Kong.

-The aims and procedures of the study were explained before written consent was obtained.

Ethical approval from the Ethics Committee of the Hong Kong Hospital Authority and the Human Subject

Ethics Subcommittee of the Hong Kong Polytechnic University was obtained prior to the start of the study.


-Analysis: 2-factor mixed repeated measures ANOVA. Intention-to-treat analysis. Dropping patients for

reasons other than the treatment were given baseline values. Dropping patients for reasons related to the

treatments were given worst score.

Participants 52 patients with chronic low-back pain (>6 months) with or without radiation. Age between 18 and 75 years.

Diagnoses: non-specific low-back pain.

Excluded: 1. Structural deformity (ankylosing spondylitis, scoliosis) 2. Lower limb fracture 3. Tumours 4.

Spinal infection 5. Cauda equina syndrome 6. Pregnancy 7. Spinal cord compression 8. Subjects who were

unable to keep the appointments 9. Receiving acupuncture treatment within the past 6 months 10. Receiving

physiotherapy treatment within the past 3 months




Previous treatments: tui na, massage, chiropractor, bone setter or corset.

Interventions 1) Electro-acupuncture: 3/week for 4 weeks by a physiotherapist certificated in acupuncture. Points were

chosen according to the literature: BL23, BL25, BL40 and SP6. Acupuncture was applied to the side on

which patients reported pain. If the reported pain was bilateral, EA was applied to the more painful side.

Sterilised disposable needles, number 30 (0.3 mm) 40-mm long needles were inserted and manipulated until

Teh Chi was obtained. Electrical stimulation on needles at a frequency of 2 Hz for 30 minutes. The intensity

of the stimulation was set at the level that the patient could tolerate and often with evoked visible muscle

contractions. The current had biphasic waveform to the four selected acupoints in two pairs. In addition, all

patients also received exercise therapy, the same as in the control group.

Randomized to this group: 26. Lost to follow-up: 1

2) Standard group exercise program led by the same physiotherapist.



The program consisted of an hourly session each week for 4 consecutive weeks, and comprised back strength-

ening and stretching exercises

In addition, patients were advised on spinal anatomy and body mechanics, back care and postural correction,

lifting and ergonomic advice, and behavioural modification, as well as a series of home exercises (15 min/day).

Randomized to this group: 26. Lost to follow-up: 2

Outcomes 1) Pain: Numerical rating scale for “average” and for “worst” pain intensity during the last week, by asking

the patient to rate perceived level of pain on a scale from 0 to 10, where 0 represents no pain and 10 represents

pain as bad as it could be.

2) Disability: The Aberdeen LBP scale (19-item) was used to measure low-back pain disability, because it is

the only LBP-specific functional disability scale validated for Chinese subjects. Responses to the questions

were summed and converted to a score percentage between 0 and 100, with 0 representing the least disabled

and 100 the most severely disabled.

These outcomes were measured immediately after, 1 month and 3 months after.

Costs: not reported

Complications: no adverse reaction or complication.





13.01

13.07

13.09

Conclusions: “Significantly better scores in the NRS and Aberdeen LBP scale were found in the exercise plus

EA group immediately after treatment, at 1-month follow-up and at 3-month follow-up”


Key:

CAM = complementary and alternative medicine

ADL = activities of daily living

WLC = waiting list control

Characteristics of excluded studies

Study Reason for exclusion

Cai 1996 Sciatica

Duplan 1983 Acute sciatica of disc origin.

Fox and Melzack 1976 Not randomized

Franke 2000 Acupuncture treatment did not involve needling.

Fujinuki 1989 Lumbar spinal canal stenosis

Galacchi 1981 Percentage of low-back pain unknown

Gallacchi 1983 Percentage of low-back pain unknown

Ghia 1976 Specific causes of low-back pain

Hackett 1988 Acupuncture treatment did not involve needling.

Ishimaru 1993 Not randomized

Junnila 1982 No back pain



Characteristics of excluded studies (Continued )

Kinoshita 1965 Sciatica



Koike 1975 Not randomized

Kuramoto 1977 Lumbar disc herniation

Laitinen 1976 Sciatica

Li 1994 Acupuncture treatment did not involve needling.

Megumi 1979 Not randomized

Ren 1996 Not randomized

Shinohara 2000 No mention of low-back pain, only musculoskeletal pain.

Sodipo 1981 Not randomized

Sugiyama 1984 Not randomized

Wang 1997 Not randomized

Wang 2000 Lumbar disc surgery

Wedenberg 2000 Pregnancy

Xingsheng 1998 Sciatica

Xu 1996 Not randomized

Yue 1978 Back (n=15) and neck (n=8) and it is a preliminary report

Zhang 1995 Not randomized

Zhang 1996 Not randomized

Zhi 1995 Not randomized

Characteristics of ongoing studies

Study Cherkin

Trial name or title Efficacy of Acupuncture for Chronic Low Back Pain

Participants Low Back Pain

Interventions Acupuncture

Outcomes

Starting date Funding: National Center for Complementary and Alternative Medicine (NCCAM)

Contact information Janet Erro, RN [email protected]

Study chairs or principal investigators

Daniel Cherkin, PhD, Study Director, Group Health Cooperative Center for Health Studies

Karen J Sherman, PhD, Principal Investigator, Group Health Cooperative Center for Health Studies

Andy Avins, MD, Principal Investigator, Kaiser Foundation Research Institute, Kaiser Permanente Northern

California

Study ID Numbers R01 AT001110-01 A1

Study Start Date April 2004

Record last reviewed March 2004

NLM Identifier NCT00065585

ClinicalTrials.gov processed this record on 2004-04-16

Notes Source: www.controlled-trials.com



Characteristics of ongoing studies (Continued )

This is a 4-arm multi-site randomized controlled trial to clarify the extent to which various types of acupuncture

needling can diminish the effect of chronic low back pain on patient functioning and symptoms. Reviews have

noted the poor quality of research in this area and urged that scientifically rigorous studies be conducted. Recent

higher quality trials suggest acupuncture is a promising treatment for back pain. This study directly addresses

methodological shortcomings that have plagued previous studies. A total of 640 subjects (160 per arm) with

low back pain lasting at least 3 months will be recruited from group model HMOs in Seattle, WA and Oakland,

CA. They will be randomized to one of three different methods of stimulation of acupuncture or to continue

usual medical care. Ten treatments will be provided over 7 weeks. The primary outcomes, dysfunction and

bothersomeness of low back pain, will be measured at baseline, and after 8, 26, and 52 weeks by telephone

interviewers masked to treatment. Analysis of covariance within an intention-to-treat context will be used to

analyze the data. Because chronic back pain is a major public health problem and the top reason patients seek

acupuncture treatment, a clear, unambiguous assessment is critical for making informed decisions about whether

acupuncture should be included as part of conventional care for back pain or covered by insurance. Results of

this study will provide the clearest evidence to date about the value of acupuncture needling as a treatment for

chronic low back pain.

Study GerAc

Trial name or title German Acupuncture Trials

Participants

Interventions

Outcomes

Starting date

Contact information http://www.gerac.de/index1.html

Notes

Study Harvard Med School

Trial name or title Physical CAM Therapies for Chronic Low Back Pain

Participants Chronic Low Back Pain

Interventions Procedure: massage therapy

Procedure: chiropractic

Procedure: acupuncture

Outcomes Study Design: Treatment, Randomized, Open Label, Active Control, Parallel Assignment

Starting date Funding: NIH

Contact information Expected Total Enrollment: 120

Location Information

Massachusetts

Harvard Vanguard Medical Associates, Boston, Massachusetts, 00000, United States

Harvard medical school, Boston, Massachusetts, 00000, United States

More Information

Study ID Numbers 1 R01 AT00622-01; EisenbergD

Study Start Date April 2002; Estimated Completion Date December 2002

Record last reviewed August 2003

NLM Identifier NCT00065975

ClinicalTrials.gov processed this record on 2004-04-16


This study compares two approaches to the management of acute low back pain: usual care (standard benefit)

vs. the choice of: usual care, chiropractic, acupuncture or massage therapy (expanded benefit). 480 subjects with

uncomplicated, acute low back pain will be recruited from a health maintenance organization, and randomized




to either usual care (n=160) or choice of expanded benefits (n=320). Patients’ preferences for individual therapies

and expectations of improvement will be measured at baseline and throughout the study. Subjects randomized

to the expanded benefits arm who choose chiropractic, acupuncture or massage will receive up to 10 treatments

over a five-week period. Additional treatments will be available after the fifth week but will require a copayment.

Treatments will be provided by licensed providers who have met strict credentialing criteria. Chiropractic,

acupuncture or massage treatments will begin within 48 hours. Chiropractic, acupuncture and massage therapy

scope of practice guidelines for the treatment of acute low back pain have been developed as have detailed

data tracking procedures to be used at each patient visit. Symptom relief, functional status, restricted activity

days, use of health care, and patient and provider satisfaction will be assessed at 2, 5,12, 26 and 52 weeks

after initiation of treatment. Primary outcomes will include: 1) change in symptoms; 2) change in functional

status; 3) patient satisfaction; and 4) total utilization of services associated with care for low back pain. Medical

records and the HMO’s cost management information system will identify use of services. It is hypothesized

that patients offered their choice of expanded benefits will experience a more rapid improvement in symptoms,

a faster return to baseline functional status, a decrease in utilization of conventional medical services, and will

be more satisfied with their care. The study is a direct examination of the effectiveness of an insurance eligibility

intervention, not a test of the efficacy of specific, non-allopathic treatment regimens. The results of this study

will provide valuable information to clinicians, patients and third party payers on the relative benefits and costs

of an “expanded benefits” treatment option which incorporates chiropractic, acupuncture and massage services

for low back

Study Kong

Trial name or title

Participants

Interventions

Outcomes

Starting date

Contact information

Notes

Study Munglani

Trial name or title Randomised controlled single-blinded trial of deep intra-muscular stimulation in the treatment of chronic

mechanical low back pain.

Participants Out-patients between 18 and 65 years old

Interventions

Outcomes

Starting date

Contact information Dr Rajesh Munglani

Contact details Box No 215

Pain Clinic

Addenbrooke’s NHS Trust

CB2 2QQ

Tel: 2346


Deep Intra-Muscular Stimulation is a technique that consists of needling the body of contracted or shortened

muscles, and it is claimed to relieve muscle spasm more effectively than other treatments, but no randomised

controlled trials have been performed, especially to compare its effects with the effects of other needling tech-

niques such as acupuncture or trigger point needling. As the Pain Clinic is at present able to make use of the

services of a highly experienced practitioner of this technique, we are planning to conduct a randomised, con-

trolled, single-blinded trial comparing the benefits of deep Intra-Muscular stimulation with superficial needling




of subcutaneous tissues in patients with chronic mechanical low back pain. We wish to asses if the needling

of deep muscles specifically produces pain relief over and above that produced by needling of more superficial

structures. We plan to treat two groups of 25 patients each, or a total of 50 patients, administering four treatment

episodes to each patient. Patients will be asked to turn up for four treatment episodes, and to fill in two self-

reporting questionnaires (SCL-90 and Pain VAS) before and at 2, 6 and 26 weeks after treatment.

Study Thomas

Trial name or title Longer term clinical and economic benefits of offering acupuncture to patients with chronic low back pain.

Participants patients with low back pain. Age 20-65 years with low back pain or sciatica, greater than 4 weeks and less than

12-months pain this episode

Interventions i) traditional Chinese acupuncture, up to 10 treatments ii) standard care offered by GP only

Outcomes

Starting date Funding: NHS

Contact information Ms Kate Thomas

Address Medical Care Research Unit

University of Sheffield

ScHARR

Regent Court

30 Regent Street

City/town Sheffield

Zip/Postcode S1 4DA

Country United Kingdom

Tel +44 0114 222 0753

Fax +44 0114 272 4095

Email [email protected]

Sponsor NHS Research and Development Health Technology Assessment Programme (HTA)


A D D I T I O N A L T A B L E S

Table 01. Methodological quality assessment

Study A and B C D, E and F G H I J K

Comments,

flaws, etc

Araki 2001 Y and Y Y Y, N, Y Y Y Y Y Y Score=10 and

no serious flaws

(High)

Carlsson 2001 Y and Y DK Y, N, Y DK DK Y (1 month);

N (3 and 6

months)

Y Y Score=7 at 1

month (follow-

up=100%),

Score=6 at 3

and 6 months

(follow-

up=64%

and 54%

respectively)

(High)



Table 01. Methodological quality assessment (Continued )


Comments,

flaws, etc

Ceccherelli

2002

Y and DK Y DK, N, Y DK DK Y Y Y Score=6. No

serious flaws.

(High)

Cherkin 2001 Y and DK Y N, N, Y Y Y Y Y Y Score=8. No

serious flaws

(High)

Coan 1980 Y and Y DK N, N, N DK N N N N Score=2 (Low)

Ding 1998 DK and N DK Y, N, N DK DK Y Y N Score=3

(Low). Main

outcome is very

subjective.

Edelist 1976 DK and DK DK Y, N, Y DK Y DK DK DK Score=3 (Low).

Main outcome

is a subjective

measure.

Methods

poorly

described.

Garvey 1989 Y and DK DK Y, N, Y Y Y Y Y Y Score=8.

No serious

flaws (High).

Baseline

characteristics

are not shown.

Groups are very

different in

size.

Giles 1999 DK and Y DK N, N, Y DK N N Y N Fatal flaw=

52% drop

out during

treatment

period in the

acupuncture

group.

Giles 2003 Y and Y Y N, N, DK Y DK N Y Y Score=6. 39%

drop out at 9-

weeks (Low).

No adjustment

for multiple

comparisons

Grant 1999 Y and Y N N, N, Y Y DK Y Y N Fatal flaw=

baseline

differences in

main outcome





Comments,

flaws, etc

measures. VAS

(range 0-200)

at baseline in

acup group was

140 and in the

TENS group

was 101.

Gunn 1980 N and DK DK N, N, DK DK DK Y N N Score=1 (Low).

Allocation by

alternation and

not concealed.

No mention

of blinded

assessments.

We don’t have

baseline values

for pain. Co-

interventions

were allowed

and not

standardized or

monitored. No

ITT: this is not

a big problem

for the 12-week

follow-up, but

maybe for the

longer term

follow-up

He 1997 DK and N Y Y, N, N DK DK N Y DK Score=3

(Low). No

information

about

allocation of

patients. No

description of

lost patients.

Inoue 2000 Y and Y DK Y,N, Y Y Y Y Y Y Score=9

(High). We

believe there

were no losses

because the

follow-up was

shortly after the

single session.

Inoue 2001 Y and Y DK Y,N,Y Y Y Y Y Y Score=9





Comments,

flaws, etc

(High). We

believe there

were no losses

because the

follow-up was

shortly after the

single session.

Kerr 2003 Y and DK DK Y,N,Y DK DK N Y N Score=4

(Low). Co-

interventions

might have

influenced the

results. Patients

followed: 76%

in the short and

66.7% in the

intermediate

follow-ups.

Kittang 2001 N and DK N DK,DK,Y Y Y Y Y Y Score=6.

No serious

flaws (High).

Baseline

differences in

three factors

(days of sick

leave previous

year, previous

attendance at

back schools

and use of pain

killers)

Kurosu 1979

(a) and (b)

DK and DK DK N, N, DK DK Y DK Y DK Score=2 (Low)

Lehmann 1983 DK and DK DK N, N, N Y DK N Y N Score=2

(Low). Follow-

up: 77%

immediately

after and 61%

after 6 months.

Leibing 2002 Y and Y Y Y,N, Y Y DK N Y DK Score=7 (High)

However, drop-

out rate: 24%

in the short

and 37% in the

long-term





Comments,

flaws, etc

Li 1997 DK and N DK Y, N, N DK DK N Y DK Score=2 (Low)

No information

about

allocation of

patients. No

description of

lost patients.

Lopacz and

Gralewski 1979

DK and DK DK N, N, N Y DK Y Y Y Score=4

(Low). No

information

about

randomization

and not

blinded.

Macdonald

1983

DK and DK Y Y, N, DK DK DK Y DK Y Score=4

(Low). No

information

about

randomization

and timing

of follow-up

measures

Mendelson

1983

DK and DK Y Y, N, Y DK DK Y Y N Score=5 (Low).

Cross over

study.

Meng 2003 Y and Y Y (pain); N

(Roland)

N, N, N Y DK Y Y Y Score=7 (small

difference in

baseline in pain

outcomes).

Score= 6

(important

baseline

difference

in RDQ

(acupuncture

group: 9.8 and

control group:

11.8). (High)

Molsberger

2002

Y and Y Y Y, N, Y DK Y Y (immed), N

(short)

Y Y Score=9

(immediately

after) and

Score=8 (short-

term: drop-

out rate at 3

months was

34%) (High).





Comments,

flaws, etc

Blinding was

between verum

and sham

acupuncture,

but not

between verum

and nothing.

Sakai 1998 DK and DK N N,N,DK DK DK N N DK Score=0 (Low).

Methods

poorly

described. A

statistically

significant

difference was

observed in

disability score

at baseline.

ADL was 7.6

in acupuncture

group and 10.3

in medication

group. Other

parameters

such as

subjective

symptom of

pain, JOA

score, duration

of pain, gender

were not

statistically

different at

baseline.

Sakai 2001 Y and Y Y N,N, Y Y Y Y Y N Score=8. No

serious flaws

(High)

Takeda &

Nabeta 2001

Y and DK DK Y,N,N DK Y Y Y DK Score=5 (Low)

Thomas and

Lundberg 1994

DK and DK Y N,N, DK N Y DK Y Y Score=4

(Low). We get

different results

when we re-

analysed using

the data from

the figures.





Comments,

flaws, etc

Tsukayama

2002

Y and Y Y N, N, Y Y Y Y Y Y Score=9.

No serious

flaws (High).

Outcome

assessor was

blinded, but

patient was

not. So it is

possible that

the blindness

was broken,

especially

because the

outcomes are

subjective.

Von Mencke

1988

DK and DK DK Y, N, Y N N N N N Score=2 (Low)

Wang 1996 DK and N DK Y, N, N DK N N DK DK Score=1 (Low).

Not adequately

randomized.

Doubts about

reliability

of outcome

measures

Wu 1991 N and N DK Y,N,N DK Y N Y DK Score=3 (Low).

Not adequately

randomized.

Doubts about

reliability

of outcome

measures

Yeung 2003 DK and Y Y N, N, Y Y Y Y Y Y Score=8.

No serious

flaws (High).

Outcome

assessor was

blinded, but

patient was

not. So it is

possible that

the blindness

was broken,

especially

because the

outcomes are

subjective. One





Comments,

flaws, etc

of the few

studies that

adjusted for

confounders in

the analysis,

but small

sample size and

did not account

for attention

effects.

Total “Yes” 17 14 14 18, 0, 19 15 15 20 28 16

Total “No” 3 5 3 15, 34, 10 2 3 12 4 10

Total “DK” 15 16 18 2, 1, 6 18 17 3 3 9

Table 02. Clinical relevance assessment

Study Patients Interventions

Relevant

outcomes Size of the effect

Benefits and

harms

Serious

deficiences?

Araki 2001 N Y Y DK DK Population is

poorly described.

Power to detect a

difference (alpha

0.05, 2-tailed) in

pain is 12% and in

function is 5.1%.

Carlsson 2001 Y Y Y Y Y

Ceccherelli 2002 Y Y N DK DK

Cherkin 2001 Y N Y DK Y Intervention is

individualized

to each patient.

Pragmatic trial.

Coan 1980 Y N Y Y DK Intervention is

poorly described

Ding 1998 Y N Y Y Y The strong and

deep needling

technique may not

be practical for

all acupuncture

settings.

Edelist 1976 N Y N N DK Irrelevant

outcomes.

Garvey 1989 N Y N Y N Benefists do not



Table 02. Clinical relevance assessment (Continued )


Relevant


Benefits and

harms

Serious

deficiences?

seem to be worth

the harms

Giles 1999 N N Y Y DK Patients and

interventions are

poorly described

Giles 2003 N Y Y DK DK Difficult to

interpret results

due to nature of

data presentation.

No follow-up

beyond 9 weeks.

Grant 1999 N N Y N N Population and

interventions are

poorly described

Gunn 1980 Y N N Y DK We don’t

know how co-

interventions were

applied. We don’t

have a separate

measure for pain.

He 1997 Y N Y Y DK No description

of acupuncture

points used.

Not sure about

validity/reliability

of outcome

measure.

Inoue 2000 N Y DK N DK

Inoue 2001 N Y DK Y Y

Kerr 2003 N Y Y N DK No clinically

important effects

detected in this

study

Kittang 2001 Y N Y N DK

Kurosu 1979 (a)

and (b)

N Y N DK DK

Li 1997 Y N Y Y DK No description

of acupuncture

points used.

Not sure about

validity/reliability

of outcome





Relevant


Benefits and

harms

Serious

deficiences?

measure.

Lehmann 1983 N N N DK Y No description

of acupuncture

points used. Teh

Chi unclear.

Leibing 2002 Y Y Y DK N

Lopacz and

Gralewski 1979

N N N DK DK Poor description

of patients and

interventions.

Macdonald 1983 Y Y Y Y DK It is not meridian

acupuncture and

the depth is too

superficial. Very

small sample size

Mendelson 1983 Y Y Y N DK

Meng 2003 Y Y Y DK DK Size of effect might

be biased by small

sample size. Harms

were assessed,

but should be

evaluated in larger

sample.

Molsberger 2002 Y Y Y Y DK

Sakai 1998 Y N Y DK DK Not sure about

validity of JOA

score. Number of

points and sessions

too small.

Sakai 2001 Y Y Y N DK Not sure about

validity of JOA

score. Number of

points and sessions

too small.

Takeda 2001 N Y Y N DK

Thomas and

Lundberg 1994

N Y Y N DK

Tsukayama 2002 Y Y Y DK N

Von Mencke 1988 Y Y Y Y DK Teh Chi unclear.

Wang 1996 Y Y Y Y Y

Wu 1991 Y Y Y Y Y





Relevant


Benefits and

harms

Serious

deficiences?

Yeung 2003 Y Y Y Y Y

Table 03. Adequacy of acupuncture

Study

Choice of

acupoints

Number of

sessions

Needling

technique Experience Control group Comments

Araki 2001 Adequate

because this is

acute low-back

pain

Adequate

because it is acute

low-back pain

Adequate Adequate Appropriate

sham

acupuncture

But there is

no description

about credibility

of sham

acupuncture.

Carlsson 2001 Adequate Adequate Adequate Adequate Adequate sham

TENS

The authors also

compared needle

acupuncture

with elec-

troacupuncture.

Ceccherelli 2002 Adequate Adequate Adequate for the

purpose of the

study, which was

to compare two

techniques of

acupuncture.

Not reported Other

acupuncture

technique

Cherkin 2001 Individualized

points.

Adequate TCM typically

with Teh Chi

Adequate Other common

therapies.

Coan 1980 Not reported Adequate Not reported Not reported Waiting list. No

treatment

Poorly reported,

but seems OK

(published in

1980).

Ding 1998 Adequate Adequate Adequate Adequate Other

acupuncture

technique

Edelist 1976 Adequate Few sessions Adequate Not reported Sham

acupuncture (but

may have some

analgesic effect)

The control

group used

needles placed

in areas devoid

of classic

acupuncture

points.

Garvey 1989

(dry needling)

Adequate (dry-

needling)

Adequate Not reported Not reported Three common

treatments

Giles 1999 Not reported Adequate Not reported. Adequate Two common

treatments:

manipulation



Table 03. Adequacy of acupuncture (Continued )

Study

Choice of

acupoints

Number of

sessions

Needling


and drugs

Giles 2003 Not reported Adequate Not reported. Adequate Two common

treatments:

manipulation

and drugs

Grant 1999 Individualized

points.

Adequate Not reported. Not reported Another

common

treatment:

TENS

Gunn 1980 (dry

needling)

Muscle motor

points. Not

adequate for dry

needling.

Adequate Adequate Not reported Standar therapy:

physiotherapy,

remedial

exercises,

occupational

therapy,

industrial

assessment.

He 1997 Adequate Adequate Adequate Not reported Chinese herbs. No information

about which

herbs were used.

Inoue 2000 Adequate Adequate for the

purpose of the

study.

Not reported Adequate Sham

acupuncture

But there is

no description

about credibility

of sham

acupuncture.

Inoue 2001 Adequate (non

meridian)

Adequate for the

purpose of the

study

Not reported Not reported Sham

acupuncture

But there is

no description

about credibility

of sham

acupuncture.

Kerr 2003 Adequate Adequate Adequate Not reported Sham TENS

Kittang 2001 Seems adequate Not reported Not reported Not reported Naproxen:

adequate dose

and duration of

treatment

Kurosu 1979 (a) Adequate Adequate for the

purpose of the

study.

Adequate Not reported Garlic

moxibustion

may be adequate

treatment for

LBP in some

cases

Kurosu 1979 (b) Adequate Adequate for the

purpose of the

Adequate Not reported Other

acupuncture




Study

Choice of

acupoints

Number of

sessions

Needling


study. technique

(needle insertion

and no retention)

Lehmann 1983 Choice of

meridians is OK

Adequate Adequate Adequate Sham TENS

Leibing 2002 Adequate Adequate Adequate Adequate Sham

acupuncture

Li 1997 Adequate Adequate Adequate Not reported Manual

acupuncture

without cupping.

Lopacz and

Gralewski 1979

Not reported Adequate Not reported Not reported Placebo: to

control for

attention effect.

Macdonald 1983 Adequate (not

meridian)

Adequate Adequate for the

purpose of the

study

Not reported Sham TENS. It is easy for

patients to

perceive that they

were receiving

different

treatments.

Mendelson 1983 Adequate Adequate Adequate Adequate Maybe not

adequate

placebo. May

have some

analgesic effect.

Meng 2003 Adequate Adequate Adequate Adequate Standard therapy

Molsberger 2002 Adequate Adequate Adequate Adequate Sham

acupuncture:

good placebo.

Sakai 1998 Adequate Adequate for the

purpose of the

study

Not reported Not reported Medication

Sakai 2001 Adequate (not

meridian)

Adequate for the

purpose of the

study

Not reported Not reported TENS: seems

adequate.

But number

of sessions too

small.

Takeda 2001 Adequate for the

purpose of the

study

Adequate Not reported Not reported Other

acupuncture

technique: local

versus distal

points.

But there is

no description

about credibility

of sham

acupuncture.

Thomas and

Lundberg 1994

Adequate Adequate Adequate Adequate No treatment

Tsukayama 2002 Adequate Adequate for the Adequate Not reported TENS but number




Study

Choice of

acupoints

Number of

sessions

Needling


purpose of the

study

of sessions too

small.

Von Mencke

1988

Adequate Adequate Adequate Not reported Sham

acupuncture

Wang 1996 Adequate Adequate Adequate Adequate Active

acupuncture:

distal points

Wu 1991 Adequate (for

acute LBP)

Adequate (single

session for acute

LBP)

Adequate Adequate Another active

acupuncture

treatment

Yeung 2003 Adequate Adequate for the

purpose of the

study

Adequate Adequate Physiotherapy

(standard

exercises)

Patients in

the exercise

group did not

receive the same

attention as in

the acupuncture

group.

Table 04. Improvement in pain

Comparison group Acute Chronic Unknown / Mixed

Acupuncture Number of studies 2 16 8

Average improvement 52% 32% 51%

Standard deviation 39% 24% 19%

Minimum 25% -17% 22%

Maximum 80% 62% 77%

No treatment Number of studies 6

Average improvement 6%

Standard deviation 25%

Minimum -33%

Maximum 42%

Sham / placebo Number of studies 1 6 3


Standard deviation 22% 17%

Minimum -19% 6%

Maximum 44% 37%

Other treatments Number of studies 1 6 3




Standard deviation 19% 73%

Minimum 0% 41%

Maximum 50% 181%

Table 05. Search Strategies

MEDLINE EMBASE

1 randomized controlled trial.pt. (72769)

2 controlled clinical trial.pt. (16977)

3 Randomized Controlled Trials/ (17706)

4 Random Allocation/ (11879)

5 Double-Blind Method/ (26902)

6 Single-Blind Method/ (4389)

7 or/1-6 (120640)

8 Animal/ not Human/ (583159)

9 7 not 8 (112795)

10 clinical trial.pt. (144571)

11 exp Clinical Trials/ (45063)

12 ((singl$ or doubl$ or trebl$ or tripl$) adj (mask$ or

blind$)).tw. (24652)

13 Placebos/ (4548)

14 placebo$.tw. (30921)

15 random$.tw. (123481)

16 Research Design/ (12824)

17 (latin adj square).tw. (663)

18 (clinic$ adj25 trial$).tw. (43883)

19 or/10-18 (275600)

20 19 not 8 (256926)

21 20 not 9 (147773)

22 Comparative Study/ (298320)

23 exp Evaluation Studies/ (155611)

24 Follow-Up Studies/ (95462)

25 Prospective Studies/ (77754)

26 (control$ or prospective$ or volunteer$).tw. (521438)

27 Cross-Over Studies/ (9791)

28 or/22-27 (917800)

29 28 not 8 (737443)

30 29 not (9 or 21) (559548)

31 9 or 21 or 30 (820116)

32 Intervertebral disk/ (1230)

33 Lumbar vertebrae/ (6673)

34 Low-back pain/ (3418)

35 Sciatica/ (544)

36 low back pain.tw. (2796)

37 backache.tw. (276)

38 lumbago.tw. (174)

39 or/32-38 (11150)

40 ACUPUNCTURE/ (114)

41 exp ACUPUNCTURE ANALGESIA/ (185)

1 clinical article/ (299265)

2 clinical study/ (2230)

3 clinical trial/ (184343)

4 controlled study/ (953915)

5 randomized controlled trial/ (58211)

6 major clinical study/ (352156)

7 double blind procedure/ (27710)

8 multicenter study/ (19950)

9 single blind procedure/ (3090)

10 crossover procedure/ (9288)

11 placebo/ (23129)

12 or/1-11 (1350338)

13 allocat$.ti,ab. (10381)

14 assign$.ti,ab. (34017)

15 blind$.ti,ab. (39706)

16 (clinic$ adj25 (study or trial)).ti,ab. (103723)

17 compar$.ti,ab. (592128)

18 control$.ti,ab. (435060)

19 cross?over.ti,ab. (7854)

20 factorial$.ti,ab. (2463)

21 follow?up.ti,ab. (3710)

22 placebo$.ti,ab. (32609)

23 prospectiv$.ti,ab. (81230)

24 random$.ti,ab. (119291)

25 ((singl$ or doubl$ or trebl$ or tripl$) adj25 (blind$ or

mask$)).ti,ab. (27455)

26 trial.ti,ab. (57298)

27 (versus or vs).ti,ab. (144304)

28 or/13-27 (1052880)

29 12 or 28 (1707121)

30 human/ (1767116)

31 nonhuman/ (855690)

32 animal/ (592)

33 animal experiment/ (266367)

34 31 or 32 or 33 (857723)

35 30 and 34 (145063)

36 29 not 34 (1118616)

37 29 and 35 (85582)

38 36 or 37 (1204198)

39 Lumbar Spine/ (4963)

40 Lumbosacral Spine/ (629)

41 Intervertebral Disk/ (710)



Table 05. Search Strategies (Continued )

MEDLINE EMBASE

42 exp ACUPUNCTURE, EAR/ (31)

43 exp ACUPUNCTURE POINTS/ (403)

44 exp ACUPUNCTURE THERAPY/ (1918)

45 acupuncture.tw. (1655)

46 electro-acupuncture.tw. (62)

47 acupressure.tw. (84)

48 or/40-47 (2324)

49 31 and 39 and 48 (49)

42 Intervertebral Disk Disease/ (295)

43 Lumbar Disk Hernia/ (835)

44 Low back pain/ (5403)

45 Ischialgia/ (743)

46 low back pain.tw. (3184)

47 backache.tw. (270)

48 lumbago.tw. (162)

49 or/39-48 (12240)

50 exp ACUPUNCTURE/ (2907)

51 exp ACUPUNCTURE ANALGESIA/ (143)

52 acupuncture.tw. (1775)

53 electro-acupuncture.tw. (56)

54 acupressure.tw. (71)

55 or/50-54 (3014)

56 38 and 49 and 55 (85)

Table 06. Internal Validity Criteria

Criteria Operationalization

A. Was the method of randomization adequate? A. A random (unpredictable) assignment sequence. Examples of

adequate methods are computer generated random number table

and use of sealed opaque envelopes. Methods of allocation using

date of birth, date of admission, hospital numbers, or alternation

should not be regarded as appropriate.

B. Was the treatment allocation concealed? B. Assignment generated by an independent person not

responsible for determining the eligibility of the patients. This

person has no information about the persons included in the

trial and has no influence on the assignment sequence or on the

decision about eligibility of the patient.

C. Were the groups similar at baseline regarding the most

important prognostic indicators?

C. In order to receive a “yes,” groups have to be similar at

baseline regarding demographic factors, duration and severity of

complaints, percentage of patients with neurologic symptoms,

and value of main outcome measure(s).

D. Was the patient blinded to the intervention? D. The reviewer determines if enough information about the

blinding is given in order to score a “yes.”

E. Was the care provider blinded to the intervention? E. The reviewer determines if enough information about the


F. Was the outcome assessor blinded to the intervention? F. The reviewer determines if enough information about the


G. Were cointerventions avoided or similar? G. Cointerventions should either be avoided in the trial design or

similar between the index and control groups.

H. Was the compliance acceptable in all groups? H. The reviewer determines if the compliance to the interventions

is acceptable, based on the reported intensity, duration, number

and frequency of sessions for both the index intervention and

control intervention(s).



Table 06. Internal Validity Criteria (Continued )

Criteria Operationalization

I. Was the drop-out rate described and acceptable? I. The number of participants who were included in the study but

did not complete the observation period or were not included in

the analysis must be described and reasons given. If the percentage

of withdrawals and drop-outs does not exceed 20% for immediate

and short-term follow-ups, 30% for intermediate and long-term

follow-ups and does not lead to substantial bias a “yes” is scored.

J. Was the timing of the outcome assessment in all groups similar? J. Timing of outcome assessment should be identical for all

intervention groups and for all important outcome assessments.

K. Did the analysis include an intention-to-treat analysis? K. All randomized patients are reported/analyzed in the group

they were allocated to by randomization for the most important

moments of effect measurement (minus missing values)

irrespective of noncompliance and cointerventions.

A N A L Y S E S

Comparison 01. acupuncture versus no treatment ((Sub)acute LBP: < 3 months)

Outcome titleNo. of

studies

No. of

participants Statistical method Effect size

01 pain Weighted Mean Difference (Fixed) 95% CI Totals not selected

02 global improvement Peto Odds Ratio 95% CI Totals not selected

03 functional status Weighted Mean Difference (Fixed) 95% CI Totals not selected

04 physical examination Weighted Mean Difference (Fixed) 95% CI Totals not selected

05 return to work Peto Odds Ratio 95% CI Totals not selected

Comparison 02. acupuncture versus placebo or sham intervention ((Sub)acute LBP: < 3 months)

Outcome titleNo. of

studies

No. of


01 pain (VAS) (lower values are

better)

Weighted Mean Difference (Random) 95% CI Totals not selected

02 global measure Peto Odds Ratio 95% CI Totals not selected

03 functional status (higher scores

are better). Generic instrument


04 physical examination: finger-

floor distance (lower values are

better)



06 mean difference in pain (final -

initial)


07 mean difference in functional

status (final - initial) Generic

instrument


08 mean difference in physical

examination (final - initial):

finger-floor distance




Comparison 03. acupuncture versus other intervention ((Sub)acute LBP: < 3 months)

Outcome titleNo. of

studies

No. of


01 pain (VAS): lower values are

better

Weighted Mean Difference (Fixed) 95% CI Totals not selected

02 global measure (higher values

are better)

Relative Risk (Random) 95% CI Totals not selected


04 physical examination (finger

floor distance)



06 Side effects / Complications Relative Risk (Random) 95% CI Totals not selected

Comparison 04. acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months)

Outcome titleNo. of

studies

No. of


01 pain Weighted Mean Difference (Fixed) 95% CI Totals not selected

02 global measure Relative Risk (Random) 95% CI Totals not selected




Comparison 05. acupuncture versus no treatment. (Chronic LBP: > 3 months)

Outcome titleNo. of

studies

No. of


01 pain (instruments: VAS and

number of words)

Standardised Mean Difference (Random) 95%

CI

Subtotals only

02 global measure (improvement) Odds Ratio (Random) 95% CI Totals not selected

03 functional status (higher values

are better)


04 limitation of activity (higher

values are worse)




07 functional status (standardized

measures)

Effect size (Random) 95% CI Subtotals only

Comparison 06. acupuncture versus placebo or sham intervention (Chronic LBP: > 3 months)

Outcome titleNo. of

studies

No. of


01 pain (lower values mean better) Weighted Mean Difference (Random) 95% CI Subtotals only

02 global improvement (higher

values are better)

Relative Risk (Random) 95% CI Subtotals only

03 pain disability index (lower

values are better)


04 physical examination

(fingertips-to-floor distance).(

Lower values are better)




05 improvement in physical

examination

Odds Ratio (Random) 95% CI Totals not selected

06 Sick leave (higher values mean

worse)

Relative Risk (Random) 95% CI Subtotals only

07 Well being (SF-36). (Higher

values are better)


08 Side effects / Complications Risk Difference (Random) 95% CI Totals not selected

09 pain (percent of baseline values) Other data No numeric data

10 sick leave Other data No numeric data

11 general level of pain (0-15

points)(more points mean less

pain)

Other data No numeric data

12 pain: difference between within

group changes

differences between (Random) 95% CI Totals not selected

13 function: difference between

within group changes


14 Pain: percentage of patients

with >50% pain reduction


15 spine range of motion:

difference between within

group changes

difference between (Random) 95% CI Totals not selected

Comparison 07. acupuncture versus other intervention. (Chronic LBP: > 3 months)

Outcome titleNo. of

studies

No. of


01 pain (lower values are better) Standardised Mean Difference (Random) 95%

CI

Totals not selected


03 back specific functional status

(lower scores mean better). Ex:

RDQ, Oswestry and Aberdeen


CI

Totals not selected


(higher scores are better). Ex:

Japan Orthopedic Association

Score.



values are better)



07 return to work (higher values

mean better)



09 pain and function (adjusted for

baseline values)


10 general level of pain (0-15

points)(more points mean less

pain)



group changes




Comparison 08. acupuncture versus acupuncture. (Chronic LBP: > 3 months)

Outcome titleNo. of

studies

No. of


01 pain (lower values mean better) Weighted Mean Difference (Fixed) 95% CI Totals not selected

02 Improvement (higher values are

better)





06 improvement Other data No numeric data

Comparison 09. dry-needling versus other intervention ((Sub)acute LBP < 3 months)

Outcome titleNo. of

studies

No. of



are better)


02 Side effects / Complications Relative Risk (Random) 95% CI Totals not selected

Comparison 10. acupuncture versus placebo or sham intervention (unknown / mixed duration of low back pain)

Outcome titleNo. of

studies

No. of


01 pain (VAS): lower values are

better



03 functional status (higher scores

are better). Generic instrument


04 physical examination (higher

values are better)



Comparison 11. acupuncture versus other intervention (unknown / mixed duration of low back pain)

Outcome titleNo. of

studies

No. of


01 pain score (lower values mean

better)


02 pain recovery: higher values are

better


CI

Totals not selected


are better)



(higher scores are better). Ex:

Japan Orthopedic Association

Score.







Comparison 12. acupuncture versus acupuncture. (unknown / mixed duration of low back pain)

Outcome titleNo. of

studies

No. of


01 pain (lower values are better) Weighted Mean Difference (Random) 95% CI Totals not selected

02 pain recovery (higher values are

better)



are better)


04 functional status (higher values

are better)


05 physical examination (finger-

floor distance) Higher values

are better.



07 improvement Other data No numeric data

Comparison 13. acupuncture plus intervention versus other intervention alone. (Chronic LBP: > 3 months)

Outcome titleNo. of

studies

No. of


01 pain (lower values are better) Standardised Mean Difference (Random) 95%

CI

Subtotals only


group changes

differences between (Random) 95% CI Subtotals only


values are better)


04 Pain: percentage of patients

with >50% pain reduction


05 function: difference between

within group changes

differences between (Random) 95% CI Subtotals only



(lower scores mean better). Ex:

RDQ, Oswestry and Aberdeen


CI

Subtotals only

08 spine range of motion:

difference between within

group changes

difference between (Random) 95% CI Totals not selected


I N D E X T E R M S

Medical Subject Headings (MeSH)∗Acupuncture Therapy; Low Back Pain [∗therapy]; Randomized Controlled Trials as Topic

MeSH check words

Humans

C O V E R S H E E T

Title Acupuncture and dry-needling for low back pain



Authors Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L, Koes BW, Berman BM

Contribution of author(s) - Furlan, van Tulder, Cherkin, Lao, Koes and Berman wrote the protocol for this review;

- Furlan, van Tulder, Koes conducted the literature search and study selection of the English

language trials;

- Tsukayama conducted the literature search and study selection of the Japanese language

trials;

- The Chinese Cochrane Centre conducted the literature search of the Chinese language

trials and Lao selected the studies;

- Furlan, van Tulder, Cherkin, and Koes performed the quality assessment and date extrac-

tion of the English language trials;

- Lao and Tsukayama performed the quality assessment and data extraction of the Japanese

and Chinese language trials;

- All authors were involved in writing the final draft of the manuscript.

Issue protocol first published 1998/2

Review first published 1999/1

Date of most recent amendment 22 August 2006

Date of most recent

SUBSTANTIVE amendment

30 October 2004

What’s New The first version of this review included 11 randomized trials. This update added 24 more

randomized trials, for a total of 35. Meta-analyses were performed for some comparison

groups.

In contrast to the previous review that concluded that the poor methodological quality of

the trials did not allow any conclusions on the effectiveness of acupuncture, the current

update demonstrated the effectiveness of meridian acupuncture for chronic low-back pain

in some special cases:

1) compared to no treatment, acupuncture improved pain and function at short-term follow-

up

2) compared to sham therapies, acupuncture improved pain at short-term follow-up, but

these effects were not maintained at longer-term follow-up and they were not observed for

functional outcomes

3) when acupuncture was added to other conventional therapies, there was better pain relief

and improved function when compared to the conventional therapies alone.

Also, this updated review examined acupuncture separately from dry-needling. The authors

concluded that no clear recommendations could be made about dry-needling because of

the small sample sizes and low methodological quality of the studies, although it appeared

that dry-needling was a useful adjunct to other therapies for chronic low-back pain. Effects

in all cases were only small.

Date new studies sought but

none found

Information not supplied by author

Date new studies found but not

yet included/excluded

Information not supplied by author

Date new studies found and

included/excluded

02 June 2003

Date authors’ conclusions

section amended

23 June 2004

Contact address Dr Andrea Furlan

Evidence-based practice co-ordinator



Institute for Work & Health

481 University Avenue, Suite 800

Toronto

ON

M5G 2E9

CANADA

E-mail: [email protected]

Tel: 416 927 2027

Fax: 416 927 4167

DOI 10.1002/14651858.CD001351.pub2

Cochrane Library number CD001351

Editorial group Cochrane Back Group

Editorial group code HM-BACK

G R A P H S A N D O T H E R T A B L E S

Figure 01. Acupuncture compared to no treatment, placebo or sham therapy



Figure 02. Acupuncture compared to another intervention or added to other interventions



Figure 03. Effects of dry-needling at trigger points

Figure 04. Comparison between two techniques of acupuncture



Analysis 01.01. Comparison 01 acupuncture versus no treatment ((Sub)acute LBP: < 3 months), Outcome 01

pain

Review: Acupuncture and dry-needling for low back pain

Comparison: 01 acupuncture versus no treatment ((Sub)acute LBP: < 3 months)

Outcome: 01 pain

Study Acupuncture No treatment Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)

N

Mean(SD) N

Mean(SD) 95% CI 95% CI

01 Immediately after the end of the sessions

02 Short-term follow-up (up to 3 months after the end of the sessions)

03 Intermediate-term follow-up (3 months to 1 year)

04 Long-term follow-up (more than 1 year)

-10 -5 0 5 10

favours acupuncture favours control




functional status



Outcome: 03 functional status


N

Mean(SD) N






-10 -5 0 5 10





physical examination



Outcome: 04 physical examination


N

Mean(SD) N






-10 -5 0 5 10




Analysis 02.01. Comparison 02 acupuncture versus placebo or sham intervention ((Sub)acute LBP: < 3

months), Outcome 01 pain (VAS) (lower values are better)


Comparison: 02 acupuncture versus placebo or sham intervention ((Sub)acute LBP: < 3 months)

Outcome: 01 pain (VAS) (lower values are better)

Study Acupuncture placebo / sham Weighted Mean Difference (Random) Weighted Mean Difference (Random)

N Mean(SD) N Mean(SD) 95% CI 95% CI

01 Immediately after end of sessions

Araki 2001 20 49.55 (22.63) 20 55.65 (27.41) -6.10 [ -21.68, 9.48 ]




-100 -50 0 50 100





months), Outcome 03 functional status (higher scores are better). Generic instrument



Outcome: 03 functional status (higher scores are better). Generic instrument




Araki 2001 20 6.60 (3.22) 20 6.50 (3.09) 0.10 [ -1.86, 2.06 ]




-4 -2 0 2 4

favours control favours acupuncture




months), Outcome 04 physical examination: finger-floor distance (lower values are better)



Outcome: 04 physical examination: finger-floor distance (lower values are better)




Araki 2001 20 25.25 (18.69) 20 28.78 (19.41) -3.53 [ -15.34, 8.28 ]




-100 -50 0 50 100





months), Outcome 06 mean difference in pain (final - initial)



Outcome: 06 mean difference in pain (final - initial)

Study Treatment Control Weighted Mean Difference (Random) Weighted Mean Difference (Random)


01 Immediately after end of sessions

Araki 2001 20 17.05 (15.74) 20 15.85 (25.50) 1.20 [ -11.93, 14.33 ]




-100 -50 0 50 100

Favours control Favours acupuncture




months), Outcome 07 mean difference in functional status (final - initial) Generic instrument



Outcome: 07 mean difference in functional status (final - initial) Generic instrument




Araki 2001 20 -2.15 (4.32) 20 -1.15 (2.68) -1.00 [ -3.23, 1.23 ]




-4 -2 0 2 4





months), Outcome 08 mean difference in physical examination (final - initial): finger-floor distance



Outcome: 08 mean difference in physical examination (final - initial): finger-floor distance




Araki 2001 20 -3.65 (7.62) 20 -0.13 (9.66) -3.52 [ -8.91, 1.87 ]




-10 -5 0 5 10




Analysis 03.01. Comparison 03 acupuncture versus other intervention ((Sub)acute LBP: < 3 months),

Outcome 01 pain (VAS): lower values are better


Comparison: 03 acupuncture versus other intervention ((Sub)acute LBP: < 3 months)

Outcome: 01 pain (VAS): lower values are better

Study Acupuncture Other intervention Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)



Kittang 2001 28 13.00 (0.0) 29 12.90 (0.0) Not estimable






-10 -5 0 5 10

favours acupunctur favours control




Outcome 02 global measure (higher values are better)



Outcome: 02 global measure (higher values are better)

Study Acupuncture Other intervention Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI





He 1997 49/50 42/50 1.17 [ 1.03, 1.33 ]

0.5 0.7 1 1.5 2

favours other interv favours acupuncture




Outcome 03 functional status





N

Mean(SD) N






-10 -5 0 5 10





Outcome 04 physical examination (finger floor distance)



Outcome: 04 physical examination (finger floor distance)










-10 -5 0 5 10





Outcome 06 Side effects / Complications



Outcome: 06 Side effects / Complications

Study Treatment Control Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI


Kittang 2001 3/28 12/29 0.26 [ 0.08, 0.82 ]




0.01 0.1 1 10 100

Favours acupuncture Favours other interv



Analysis 04.01. Comparison 04 acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months), Outcome 01

pain


Comparison: 04 acupuncture versus acupuncture. ((Sub)acute LBP: < 3 months)

Outcome: 01 pain

Study Technique 1 Technique 2 Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)

N

Mean(SD) N






-10 -5 0 5 10





global measure



Outcome: 02 global measure

Study Technique 1 Technique 2 Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI


Wu (a) 1991 70/75 60/75 1.17 [ 1.03, 1.33 ]

Wu (b) 1991 60/75 45/75 1.33 [ 1.07, 1.66 ]




0.5 0.7 1 1.5 2

favours technique 2 favours technique 1




functional status





N

Mean(SD) N






-10 -5 0 5 10










N

Mean(SD) N






-10 -5 0 5 10




Analysis 05.01. Comparison 05 acupuncture versus no treatment. (Chronic LBP: > 3 months), Outcome 01

pain (instruments: VAS and number of words)


Comparison: 05 acupuncture versus no treatment. (Chronic LBP: > 3 months)

Outcome: 01 pain (instruments: VAS and number of words)

Study Acupuncture No treatment Standardised Mean Difference (Random) Weight Standardised Mean Difference (Random)

N Mean(SD) N Mean(SD) 95% CI (%) 95% CI


Subtotal (95% CI) 0 0 0.0 Not estimable

Test for heterogeneity: not applicable

Test for overall effect: not applicable


Coan 1980 25 2.84 (2.00) 25 4.69 (2.00) 60.5 -0.91 [ -1.50, -0.33 ]

Thomas % Lundberg 94 30 4.00 (5.00) 10 6.10 (1.75) 39.5 -0.46 [ -1.19, 0.26 ]

Subtotal (95% CI) 55 35 100.0 -0.73 [ -1.19, -0.28 ]

Test for heterogeneity chi-square=0.89 df=1 p=0.35 I² =0.0%

Test for overall effect z=3.16 p=0.002


Thomas % Lundberg 94 30 4.00 (3.00) 10 6.20 (1.80) 100.0 -0.78 [ -1.52, -0.04 ]

Subtotal (95% CI) 30 10 100.0 -0.78 [ -1.52, -0.04 ]







-4 -2 0 2 4

favours acupuncture favours no treatment




global measure (improvement)



Outcome: 02 global measure (improvement)

Study Acupuncture No treatment Odds Ratio (Random) Odds Ratio (Random)

n/N n/N 95% CI 95% CI



Coan 1980 19/25 5/25 12.67 [ 3.31, 48.50 ]



0.01 0.1 1 10 100

favours no treatment favours acupuncture




functional status (higher values are better)



Outcome: 03 functional status (higher values are better)

Study Acupuncture No treatment Weighted Mean Difference (Random) Weighted Mean Difference (Random)




Thomas % Lundberg 94 30 8.50 (2.30) 10 7.00 (2.50) 1.50 [ -0.25, 3.25 ]


Thomas % Lundberg 94 30 8.90 (3.10) 10 9.00 (2.40) -0.10 [ -1.96, 1.76 ]


-4 -2 0 2 4

favours no treatment favours acupuncture




limitation of activity (higher values are worse)



Outcome: 04 limitation of activity (higher values are worse)





Coan 1980 25 1.33 (0.70) 25 1.77 (0.70) -0.44 [ -0.83, -0.05 ]



-1 -0.5 0 0.5 1










N

Mean(SD) N






-10 -5 0 5 10





functional status (standardized measures)



Outcome: 07 functional status (standardized measures)

Study Acupuncture No treatment Effect size (SE) Effect size (Random) Weight Effect size (Random)

N N 95% CI (%) 95% CI


Subtotal (95% CI) 0.0 Not estimable




Coan 1980 25 25 0.63 (0.29) 61.9 0.63 [ 0.06, 1.20 ]

Thomas % Lundberg 94 30 10 0.64 (0.37) 38.1 0.64 [ -0.09, 1.37 ]

Subtotal (95% CI) 100.0 0.63 [ 0.19, 1.08 ]




Thomas % Lundberg 94 30 10 0.03 (0.37) 100.0 0.03 [ -0.70, 0.76 ]

Subtotal (95% CI) 100.0 0.03 [ -0.70, 0.76 ]







-4 -2 0 2 4

Favours no treatment Favours acupuncture



Analysis 06.01. Comparison 06 acupuncture versus placebo or sham intervention (Chronic LBP: > 3 months),

Outcome 01 pain (lower values mean better)


Comparison: 06 acupuncture versus placebo or sham intervention (Chronic LBP: > 3 months)

Outcome: 01 pain (lower values mean better)

Study Acupuncture placebo / sham Weighted Mean Difference (Random) Weight Weighted Mean Difference (Random)



Kerr 2003 26 51.30 (22.40) 20 61.70 (30.60) 9.0 -10.40 [ -26.34, 5.54 ]

Leibing 2002 35 21.00 (22.00) 40 32.00 (22.00) 22.9 -11.00 [ -20.98, -1.02 ]

Mendelson 1983 36 30.20 (18.00) 41 40.00 (24.30) 25.3 -9.80 [ -19.28, -0.32 ]

Molsberger 2002 58 26.00 (21.00) 58 36.00 (19.00) 42.9 -10.00 [ -17.29, -2.71 ]

Subtotal (95% CI) 155 159 100.0 -10.21 [ -14.99, -5.44 ]




Carlsson (even) 34 52.00 (24.00) 16 64.00 (25.00) 27.7 -12.00 [ -26.67, 2.67 ]

Molsberger 2002 47 23.00 (20.00) 41 43.00 (23.00) 72.3 -20.00 [ -29.07, -10.93 ]

Subtotal (95% CI) 81 57 100.0 -17.79 [ -25.50, -10.07 ]


Test for overall effect z=4.52 p<0.00001


Carlsson (even) 23 48.00 (22.00) 9 62.00 (30.00) 17.4 -14.00 [ -35.56, 7.56 ]

Leibing 2002 33 31.00 (18.00) 31 35.00 (22.00) 82.6 -4.00 [ -13.88, 5.88 ]

Subtotal (95% CI) 56 40 100.0 -5.74 [ -14.72, 3.25 ]




Carlsson (even) 21 42.00 (24.00) 6 54.00 (35.00) 100.0 -12.00 [ -41.83, 17.83 ]

Subtotal (95% CI) 21 6 100.0 -12.00 [ -41.83, 17.83 ]



-100 -50 0 50 100





Outcome 02 global improvement (higher values are better)



Outcome: 02 global improvement (higher values are better)

Study Acupuncture placebo / sham Relative Risk (Random) Weight Relative Risk (Random)

n/N n/N 95% CI (%) 95% CI


Lopacz % Gralewski 13/18 9/16 11.2 1.28 [ 0.76, 2.16 ]

Mendelson 1983 21/36 21/41 18.2 1.14 [ 0.76, 1.71 ]

Molsberger 2002 52/62 41/61 70.6 1.25 [ 1.02, 1.53 ]

Subtotal (95% CI) 116 118 100.0 1.23 [ 1.04, 1.46 ]

Total events: 86 (Acupuncture), 71 (placebo / sham)




Carlsson 2001 16/34 2/16 9.8 3.76 [ 0.98, 14.44 ]

Edelist 1976 7/15 6/15 22.7 1.17 [ 0.51, 2.66 ]

Molsberger 2002 36/49 23/42 67.5 1.34 [ 0.97, 1.85 ]

Subtotal (95% CI) 98 73 100.0 1.44 [ 0.92, 2.24 ]





Kerr 2003 21/23 13/17 100.0 1.19 [ 0.89, 1.60 ]

Subtotal (95% CI) 23 17 100.0 1.19 [ 0.89, 1.60 ]





Carlsson 2001 14/34 2/16 100.0 3.29 [ 0.85, 12.80 ]

Subtotal (95% CI) 34 16 100.0 3.29 [ 0.85, 12.80 ]




0.1 0.2 0.5 1 2 5 10





Outcome 03 pain disability index (lower values are better)



Outcome: 03 pain disability index (lower values are better)




Leibing 2002 35 11.30 (15.00) 40 15.80 (10.50) -4.50 [ -10.44, 1.44 ]



Leibing 2002 33 16.20 (12.50) 31 17.00 (11.30) -0.80 [ -6.63, 5.03 ]


-10 -5 0 5 10





Outcome 04 physical examination (fingertips-to-floor distance).( Lower values are better)



Outcome: 04 physical examination (fingertips-to-floor distance).( Lower values are better)

Study Acupuncture placebo / sham Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)



Kerr 2003 26 23.60 (20.00) 20 28.40 (24.20) -4.80 [ -17.90, 8.30 ]




-100 -50 0 50 100





Outcome 05 improvement in physical examination



Outcome: 05 improvement in physical examination

Study Treatment Control Odds Ratio (Random) Odds Ratio (Random)

n/N n/N 95% CI 95% CI



Edelist 1976 6/15 5/15 1.33 [ 0.30, 5.91 ]



0.1 0.2 0.5 1 2 5 10





Outcome 06 Sick leave (higher values mean worse)



Outcome: 06 Sick leave (higher values mean worse)

Study Acupuncture placebo / sham Relative Risk (Random) Weight Relative Risk (Random)

n/N n/N 95% CI (%) 95% CI












Carlsson 2001 3/21 5/11 38.5 0.31 [ 0.09, 1.08 ]

Lehmann 1986 6/13 7/13 61.5 0.86 [ 0.40, 1.86 ]

Subtotal (95% CI) 34 24 100.0 0.58 [ 0.22, 1.54 ]









0.1 0.2 0.5 1 2 5 10





Outcome 07 Well being (SF-36). (Higher values are better)



Outcome: 07 Well being (SF-36). (Higher values are better)

Study Treatment Control Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)



Kerr 2003 26 63.90 (20.30) 20 57.50 (23.20) 6.40 [ -6.42, 19.22 ]




-100 -50 0 50 100









Study Treatment Control Risk Difference (Random) Risk Difference (Random)

n/N n/N 95% CI 95% CI


Carlsson 2001 0/34 0/16 0.0 [ -0.09, 0.09 ]

Leibing 2002 3/35 0/40 0.09 [ -0.02, 0.19 ]



Carlsson 2001 0/21 0/6 0.0 [ -0.20, 0.20 ]

Lehmann 1986 0/13 0/13 0.0 [ -0.14, 0.14 ]


-0.5 -0.25 0 0.25 0.5



Outcome 09 pain (percent of baseline values)

Immediately after the end of the sessionsStudy

Short-term follow-up (up to 3 months after the end of the sessions)Study Group Follow-up Number of patients Percent of baseline Standard Deviation p value

Carlsson (even) Acupuncture 1 month 34 87% 32 0.003

Carlsson (even) Placebo 1 month 16 123% 46

Carlsson (morn) Acupuncture 1 month 34 88% 32 0.000

Carlsson (morn) Placebo 1 month 16 138% 40

Intermediate-term follow-up (3 months to 1 year)Study Group Follow-up Number of patients Percent of baseline Standard Deviation p value

Carlsson (even) Acupuncture 3 months 23 75% 34 0.007

Carlsson (even) Placebo 3 months 9 120% 50

Carlsson (morn) Acupuncture 3 months 23 76% 37 0.001

Carlsson (morn) Placebo 3 months 9 130% 39



Long-term follow-up (more than 1 year)

Study Group Follow-up

Number of

patients Percent of baseline

Standard

Deviation p value

Carlsson (even) Acupuncture 6 months or longer 21 69% 31 0.056

Carlsson (even) Placebo 6 months or longer 6 100% 48

Carlsson (morn) Acupuncture 6 months or longer 21 76% 33 0.128

Carlsson (morn) Placebo 6 months or longer 6 133% 76


Outcome 10 sick leave

Long-term follow-upStudy Group Time Full time work Sick leave part-time Sick leave full time p value

Carlsson 2001 Acupuncture Baseline 7 6 8

Carlsson 2001 After 6 months 11 7 3 0.024

Carlsson 2001 Placebo Baseline 4 2 5

Carlsson 2001 After 6 months 5 1 5 0.655


Outcome 11 general level of pain (0-15 points)(more points mean less pain)

Immediately after the end of the sessionsStudy Group Value p value

Lehmann 1986 Acupuncture 10.59 F 2,50 = 1.66 (p<0.2)

Lehmann 1986 TENS 9.16

Lehmann 1986 Sham TENS 9.00

Short-term follow-up (up to 3 months after the end of the sessions)Study

Intermediate-term follow-up (3 months to 1 year)Study Group Value p value

Lehmann 1986 Acupuncture 11.08 F 2,41=3.57

p=0.04 (not adjusted for multiple comparisons).

p=0.1 (adjusted for multiple comparisons)





Long-term follow-up (more than 1 year)Study


Outcome 12 pain: difference between within group changes



Outcome: 12 pain: difference between within group changes

Study differences between (SE) differences between (Random) differences between (Random)

95% CI 95% CI


Leibing 2002 -0.60 (0.54) -0.60 [ -1.66, 0.46 ]



Leibing 2002 -0.10 (0.53) -0.10 [ -1.14, 0.94 ]


-4 -2 0 2 4

Favours acupuncture Favours control


Outcome 13 function: difference between within group changes



Outcome: 13 function: difference between within group changes


95% CI 95% CI


Leibing 2002 -4.20 (2.96) -4.20 [ -10.00, 1.60 ]



Leibing 2002 -0.50 (2.98) -0.50 [ -6.34, 5.34 ]


-10 -5 0 5 10





Outcome 14 Pain: percentage of patients with >50% pain reduction



Outcome: 14 Pain: percentage of patients with >50% pain reduction

Study Treatment Control Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI


Molsberger 2002 39/60 20/58 1.89 [ 1.26, 2.81 ]


Molsberger 2002 36/47 12/41 2.62 [ 1.59, 4.32 ]



0.1 0.2 0.5 1 2 5 10

Favours sham Favours acupuncture


Outcome 15 spine range of motion: difference between within group changes



Outcome: 15 spine range of motion: difference between within group changes

Study difference between (SE) difference between (Random) difference between (Random)

95% CI 95% CI


Leibing 2002 -3.50 (3.26) -3.50 [ -9.89, 2.89 ]



Leibing 2002 -0.90 (3.19) -0.90 [ -7.15, 5.35 ]


-10 -5 0 5 10

Favours acupuncture Favours sham



Analysis 07.01. Comparison 07 acupuncture versus other intervention. (Chronic LBP: > 3 months), Outcome

01 pain (lower values are better)


Comparison: 07 acupuncture versus other intervention. (Chronic LBP: > 3 months)

Outcome: 01 pain (lower values are better)

Study Acupuncture Other intervention Standardised Mean Difference (Random) Standardised Mean Difference (Random)



Giles 1999 (manip) 16 5.10 (7.80) 32 2.50 (6.96) 0.35 [ -0.25, 0.96 ]

Giles 1999 (NSAID) 16 5.10 (7.80) 20 3.80 (4.81) 0.20 [ -0.46, 0.86 ]

Giles 2003 (manip) 33 7.00 (5.18) 35 3.00 (5.18) 0.76 [ 0.27, 1.26 ]

Giles 2003 (NSAID) 33 7.00 (5.18) 39 5.00 (3.70) 0.45 [ -0.02, 0.92 ]

Grant 1999 32 0.0 (0.0) 28 0.0 (0.0) Not estimable


Cherkin 2001 (mass) 94 4.00 (0.40) 78 3.60 (0.30) 1.11 [ 0.79, 1.43 ]

Cherkin 2001 (sc) 94 4.00 (0.40) 90 4.60 (0.40) -1.49 [ -1.82, -1.17 ]


Cherkin 2001 (mass) 94 4.50 (0.40) 78 3.20 (0.40) 3.24 [ 2.78, 3.69 ]

Cherkin 2001 (sc) 94 4.50 (0.40) 90 3.80 (0.40) 1.74 [ 1.40, 2.08 ]


-4 -2 0 2 4

favours acupunctur favours other interv




03 back specific functional status (lower scores mean better). Ex: RDQ, Oswestry and Aberdeen



Outcome: 03 back specific functional status (lower scores mean better). Ex: RDQ, Oswestry and Aberdeen




Giles 1999 (manip) 16 24.50 (26.60) 32 19.50 (30.74) 0.17 [ -0.43, 0.77 ]

Giles 1999 (NSAID) 16 24.50 (26.60) 20 20.00 (21.47) 0.18 [ -0.47, 0.84 ]

Giles 2003 (manip) 34 26.00 (20.70) 35 14.00 (24.40) 0.52 [ 0.04, 1.00 ]

Giles 2003 (NSAID) 34 26.00 (20.70) 40 32.00 (23.70) -0.27 [ -0.72, 0.19 ]


Cherkin 2001 (mass) 94 7.90 (0.70) 78 6.30 (0.60) 2.43 [ 2.03, 2.82 ]

Cherkin 2001 (sc) 94 7.90 (0.70) 90 8.80 (0.70) -1.28 [ -1.60, -0.96 ]


Cherkin 2001 (mass) 94 8.00 (0.70) 78 6.80 (0.70) 1.71 [ 1.36, 2.06 ]

Cherkin 2001 (sc) 94 8.00 (0.70) 90 6.40 (0.70) 2.28 [ 1.90, 2.65 ]


-4 -2 0 2 4

favours acupunctur favours other interv


04 back specific functional status (higher scores are better). Ex: Japan Orthopedic Association Score.



Outcome: 04 back specific functional status (higher scores are better). Ex: Japan Orthopedic Association Score.

Study Acupuncture Other intervention Weighted Mean Difference (Random) Weighted Mean Difference (Random)

N

Mean(SD) N






-4 -2 0 2 4

favours acupuncture favours other interv




05 pain disability index (lower values are better)




Study Treatment Control Weighted Mean Difference (Random) Weighted Mean Difference (Random)

N

Mean(SD) N






-100 -50 0 50 100



06 physical examination





N

Mean(SD) N






-10 -5 0 5 10





07 return to work (higher values mean better)



Outcome: 07 return to work (higher values mean better)


n/N n/N 95% CI 95% CI




Lehmann 1986 7/13 8/13 0.88 [ 0.45, 1.70 ]


0.1 0.2 0.5 1 2 5 10



08 Side effects / Complications





n/N n/N 95% CI 95% CI


Giles 1999 (manip) 0/16 0/32 0.0 [ -0.09, 0.09 ]

Giles 1999 (NSAID) 0/16 3/20 -0.15 [ -0.33, 0.03 ]

Grant 1999 3/32 3/28 -0.01 [ -0.17, 0.14 ]

Tsukayama 2002 4/9 3/10 0.14 [ -0.29, 0.58 ]


Cherkin 2001 (mass) 0/94 0/78 0.0 [ -0.02, 0.02 ]

Cherkin 2001 (sc) 0/94 0/90 0.0 [ -0.02, 0.02 ]


Cherkin 2001 (mass) 0/94 0/78 0.0 [ -0.02, 0.02 ]

Cherkin 2001 (sc) 0/94 0/90 0.0 [ -0.02, 0.02 ]

Lehmann 1986 0/13 0/13 0.0 [ -0.14, 0.14 ]


-1 -0.5 0 0.5 1





09 pain and function (adjusted for baseline values)

Immediately after the end of teh sessionsStudy Comparison Outcome measure Timing p value

Tsukayama 2002 Acupuncture versus TENS Functional status (JOA): higher scores are better Immediately after 0.24

Tsukayama 2002

Short-term follow-up (up to 3 months after the end of the sessions)Study Comparison Outcome measure Timing p value

Cherkin 2001 (mass) Acupuncture versus massage Pain 9 weeks 0.23

Cherkin 2001 (mass) Function 9 weeks 0.01 (massage is better)

Cherkin 2001 (sc) Acupuncture versus self-care education Pain 9 weeks 0.55

Cherkin 2001 (sc) Function 9 weeks 0.75

Intermediate-term follow-up (3 months to 1 year)Study Comparison Outcome measure Timing p value

Cherkin 2001 (mass) Acupuncture versus massage Pain 52 weeks 0.002 (massage is better)

Cherkin 2001 (mass) Function 52 weeks 0.05 (massage is better)

Cherkin 2001 (sc) Acupuncture versus self-care education Pain 52 weeks 0.10

Cherkin 2001 (sc) Function 52 weeks 0.10



10 general level of pain (0-15 points)(more points mean less pain)

Immediately after the end of the sessionsStudy Group Value p value

Lehmann 1986 Acupuncture 10.59 F 2,50 = 1.66 (p<0.2)






Intermediate-term follow-up (3 months to 1 year)Study Group Value p value

Lehmann 1986 Acupuncture 11.08 F 2,41=3.57

p=0.04 (not adjusted for multiple comparisons).

p=0.1 (adjusted for multiple comparisons)





11 pain: difference between within group changes





95% CI 95% CI


Tsukayama 2002 -2.10 (0.86) -2.10 [ -3.79, -0.41 ]




-4 -2 0 2 4




Analysis 08.01. Comparison 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months), Outcome 01

pain (lower values mean better)


Comparison: 08 acupuncture versus acupuncture. (Chronic LBP: > 3 months)

Outcome: 01 pain (lower values mean better)




Ceccherelli 2002 21 14.54 (10.88) 21 22.25 (16.08) -7.71 [ -16.01, 0.59 ]


Ceccherelli 2002 21 7.50 (12.94) 21 18.00 (17.16) -10.50 [ -19.69, -1.31 ]



-100 -50 0 50 100



Improvement (higher values are better)



Outcome: 02 Improvement (higher values are better)


n/N n/N 95% CI 95% CI



Carlsson 2001 8/18 7/16 1.02 [ 0.48, 2.17 ]


Carlsson 2001 8/19 6/16 1.12 [ 0.49, 2.56 ]


0.2 0.5 1 2 5

favours Technique 2 favours Technique 1




functional status




Study Acupuncture Conventional tx Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)

N

Mean(SD) N






-10 -5 0 5 10







Study Acupuncture Conventional tx Weighted Mean Difference (Fixed) Weighted Mean Difference (Fixed)

N

Mean(SD) N






-10 -5 0 5 10



improvement

Immediately after the end of the sessionsStudy Improvement Technique 1: Regular Technique 2: Ancient p value

Ding 1998 Marked effective 4 8

Ding 1998 Improved 6 3

Ding 1998 No change 5 2



Immediately after the end of the sessions (Continued )

Study Improvement Technique 1: Regular Technique 2: Ancient p value

Short-term follow-up (up to 3 months after the end of the sessions)Study Improvement Technique 1: Regular Technique 2: Ancient p value

Ding 1998 Cure 4 22 Chi-square=12.44

p<0.01

Ding 1998 Marked effective 4 8

Ding 1998 Improved 6 3

Ding 1998 No change 5 2

Intermediate-term follow-up (3 months to 1 year)Study


Analysis 09.01. Comparison 09 dry-needling versus other intervention ((Sub)acute LBP < 3 months),

Outcome 01 global measure (higher values are better)


Comparison: 09 dry-needling versus other intervention ((Sub)acute LBP < 3 months)


Study Dry-needling Other intervention Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI


Gunn 1980 28/29 16/27 1.63 [ 1.18, 2.24 ]


Garvey 1989 (lidoc) 11/20 4/13 1.79 [ 0.72, 4.43 ]

Garvey 1989 (spray) 11/20 8/16 1.10 [ 0.59, 2.07 ]

Garvey 1989(steroid) 11/20 5/14 1.54 [ 0.69, 3.45 ]

Gunn 1980 27/29 14/27 1.80 [ 1.23, 2.62 ]


Gunn 1980 21/21 14/23 1.64 [ 1.18, 2.28 ]


0.1 0.2 0.5 1 2 5 10

Favours control Favours dry-needling



Analysis 09.02. Comparison 09 dry-needling versus other intervention ((Sub)acute LBP < 3 months),



Comparison: 09 dry-needling versus other intervention ((Sub)acute LBP < 3 months)


Study Dry-needling Control Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI



Garvey 1989 3/20 1/14 2.10 [ 0.24, 18.17 ]



0.01 0.1 1 10 100

Favours dry-needling Favours other interv

Analysis 10.01. Comparison 10 acupuncture versus placebo or sham intervention (unknown / mixed duration

of low back pain), Outcome 01 pain (VAS): lower values are better


Comparison: 10 acupuncture versus placebo or sham intervention (unknown / mixed duration of low back pain)

Outcome: 01 pain (VAS): lower values are better



01 Short term (immediately after end of sessions)

Inoue 2000 15 3.90 (2.60) 12 3.60 (2.10) 0.30 [ -1.47, 2.07 ]

Inoue 2001 10 3.73 (2.44) 11 6.41 (1.35) -2.68 [ -4.39, -0.97 ]




-10 -5 0 5 10





of low back pain), Outcome 02 global measure




Study Acupuncture placebo / sham Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI



Von Mencke 1988 33/35 15/30 1.89 [ 1.31, 2.72 ]



0.1 0.2 0.5 1 2 5 10

favours placebo/sham favours acupuncture


of low back pain), Outcome 03 functional status (higher scores are better). Generic instrument



Outcome: 03 functional status (higher scores are better). Generic instrument


N

Mean(SD) N






-4 -2 0 2 4





of low back pain), Outcome 04 physical examination (higher values are better)



Outcome: 04 physical examination (higher values are better)


N

Mean(SD) N






-10 -5 0 5 10


Analysis 11.01. Comparison 11 acupuncture versus other intervention (unknown / mixed duration of low back

pain), Outcome 01 pain score (lower values mean better)


Comparison: 11 acupuncture versus other intervention (unknown / mixed duration of low back pain)

Outcome: 01 pain score (lower values mean better)




Sakai 1998 14 2.30 (1.50) 12 2.40 (1.50) -0.10 [ -1.26, 1.06 ]

Sakai 2001 31 5.30 (3.00) 33 5.90 (3.40) -0.60 [ -2.17, 0.97 ]




-4 -2 0 2 4

favours acupuncture favours other interv




pain), Outcome 02 pain recovery: higher values are better



Outcome: 02 pain recovery: higher values are better




Kurosu 1979(a) 10 0.86 (0.80) 10 1.18 (0.75) -0.40 [ -1.28, 0.49 ]




-4 -2 0 2 4



pain), Outcome 03 global measure (higher values are better)





n/N n/N 95% CI 95% CI


Sakai 2001 13/31 10/33 1.38 [ 0.71, 2.69 ]




0.2 0.5 1 2 5





pain), Outcome 04 back specific functional status (higher scores are better). Ex: Japan Orthopedic Association

Score.



Outcome: 04 back specific functional status (higher scores are better). Ex: Japan Orthopedic Association Score.




Sakai 1998 14 12.10 (2.00) 12 13.30 (0.80) -1.20 [ -2.34, -0.06 ]

Sakai 2001 31 15.90 (2.00) 33 15.80 (2.60) 0.10 [ -1.03, 1.23 ]




-10 -5 0 5 10



pain), Outcome 05 physical examination





N

Mean(SD) N






-10 -5 0 5 10





pain), Outcome 07 Side effects / Complications





n/N n/N 95% CI 95% CI


Sakai 2001 0/31 2/33 -0.06 [ -0.16, 0.04 ]




-0.5 -0.25 0 0.25 0.5


Analysis 12.01. Comparison 12 acupuncture versus acupuncture. (unknown / mixed duration of low back

pain), Outcome 01 pain (lower values are better)


Comparison: 12 acupuncture versus acupuncture. (unknown / mixed duration of low back pain)


Study Technique 1 Technique 2 Weighted Mean Difference (Random) Weighted Mean Difference (Random)


01 Short term (immediately after end of sessions)

Takeda % Nabeta 2001 9 28.00 (24.30) 9 17.00 (20.90) 11.00 [ -9.94, 31.94 ]




-100 -50 0 50 100





pain), Outcome 02 pain recovery (higher values are better)



Outcome: 02 pain recovery (higher values are better)

Study Techique 1 Technique 2 Weighted Mean Difference (Random) Weighted Mean Difference (Random)



Kurosu 1979(b) 10 1.10 (0.83) 10 0.43 (0.39) 0.67 [ 0.10, 1.24 ]




-4 -2 0 2 4



pain), Outcome 03 global measure (higher values are better)





n/N n/N 95% CI 95% CI



Wang 1996 239/246 175/246 1.37 [ 1.26, 1.48 ]



0.5 0.7 1 1.5 2





pain), Outcome 04 functional status (higher values are better)



Outcome: 04 functional status (higher values are better)




Takeda % Nabeta 2001 9 14.40 (1.60) 9 15.00 (1.40) -0.60 [ -1.99, 0.79 ]




-10 -5 0 5 10



pain), Outcome 05 physical examination (finger-floor distance) Higher values are better.



Outcome: 05 physical examination (finger-floor distance) Higher values are better.




Takeda % Nabeta 2001 9 14.30 (9.90) 9 18.00 (10.00) -3.70 [ -12.89, 5.49 ]




-10 -5 0 5 10



pain), Outcome 07 improvement

Immediately after the end of the sessionsStudy




Intermediate-term follow-up (3 months to 1 year)Study

Long-term follow-up (more than 1 year)Study Improvement Technique 1: Ac+cupp Technique 2: Acup p value

Li & Shang 1997 Cure 33 22 <0.01

Li & Shang 1997 Marked effective 32 28

Li & Shang 1997 Improved 13 26

Li & Shang 1997 No change 0 2

Analysis 13.01. Comparison 13 acupuncture plus intervention versus other intervention alone. (Chronic LBP:

> 3 months), Outcome 01 pain (lower values are better)


Comparison: 13 acupuncture plus intervention versus other intervention alone. (Chronic LBP: > 3 months)


Study Acup + intervention Intervention alone Standardised Mean Difference (Random) Weight Standardised Mean Difference (Random)



Leibing 2002 35 2.10 (2.20) 39 4.40 (1.70) 24.1 -1.17 [ -1.66, -0.67 ]

Meng 2003 24 1.50 (1.20) 23 2.40 (1.30) 17.7 -0.71 [ -1.30, -0.12 ]

Molsberger 2002 58 2.60 (2.10) 58 3.90 (2.10) 38.5 -0.61 [ -0.99, -0.24 ]

Yeung 2003 26 3.81 (2.10) 26 5.12 (2.18) 19.7 -0.60 [ -1.16, -0.05 ]

Subtotal (95% CI) 143 146 100.0 -0.76 [ -1.02, -0.50 ]




Meng 2003 24 1.30 (0.80) 23 2.40 (1.00) 30.3 -1.20 [ -1.82, -0.57 ]

Molsberger 2002 47 2.30 (2.00) 36 5.20 (1.90) 36.4 -1.47 [ -1.96, -0.98 ]

Yeung 2003 26 3.77 (2.12) 26 5.19 (2.47) 33.3 -0.61 [ -1.16, -0.05 ]

Subtotal (95% CI) 97 85 100.0 -1.10 [ -1.62, -0.58 ]


-4 -2 0 2 4

favours acup + inter favours inter alone (Continued . . . )



(. . . Continued)





Leibing 2002 33 3.10 (1.80) 30 4.50 (2.00) 55.1 -0.73 [ -1.24, -0.22 ]

Yeung 2003 26 3.46 (2.18) 26 5.27 (2.31) 44.9 -0.79 [ -1.36, -0.23 ]

Subtotal (95% CI) 59 56 100.0 -0.76 [ -1.14, -0.38 ]







-4 -2 0 2 4

favours acup + inter favours inter alone


> 3 months), Outcome 02 pain: difference between within group changes




Study differences between (SE) differences between (Random) Weight differences between (Random)

95% CI (%) 95% CI


Leibing 2002 -1.70 (0.53) 42.9 -1.70 [ -2.74, -0.66 ]

Meng 2003 -0.60 (0.33) 57.1 -0.60 [ -1.25, 0.05 ]

Subtotal (95% CI) 100.0 -0.91 [ -1.46, -0.36 ]




Meng 2003 -0.70 (0.32) 100.0 -0.70 [ -1.33, -0.07 ]

Subtotal (95% CI) 100.0 -0.70 [ -1.33, -0.07 ]




Leibing 2002 -0.80 (0.51) 100.0 -0.80 [ -1.80, 0.20 ]

-4 -2 0 2 4

Favours acup + inter Favours interv alone (Continued . . . )



(. . . Continued)


95% CI (%) 95% CI

Subtotal (95% CI) 100.0 -0.80 [ -1.80, 0.20 ]







-4 -2 0 2 4

Favours acup + inter Favours interv alone


> 3 months), Outcome 03 pain disability index (lower values are better)




Study Acup + interv Interv alone Weighted Mean Difference (Random) Weighted Mean Difference (Random)



Leibing 2002 40 11.30 (15.00) 46 22.30 (7.80) -11.00 [ -16.17, -5.83 ]



Leibing 2002 40 16.20 (12.50) 46 22.60 (10.00) -6.40 [ -11.23, -1.57 ]


-100 -50 0 50 100





> 3 months), Outcome 04 Pain: percentage of patients with >50% pain reduction



Outcome: 04 Pain: percentage of patients with >50% pain reduction

Study Acup + intervention Intervention alone Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI


Molsberger 2002 39/60 23/53 1.50 [ 1.05, 2.15 ]


Molsberger 2002 36/47 5/36 5.51 [ 2.41, 12.63 ]



0.1 0.2 0.5 1 2 5 10

Favours interv alone Favours acup + inter


> 3 months), Outcome 05 function: difference between within group changes



Outcome: 05 function: difference between within group changes


95% CI (%) 95% CI


Leibing 2002 -11.30 (2.95) 44.9 -11.30 [ -17.08, -5.52 ]

Meng 2003 -2.60 (0.98) 55.1 -2.60 [ -4.52, -0.68 ]

Subtotal (95% CI) 100.0 -3.46 [ -5.29, -1.64 ]




Meng 2003 -3.10 (1.10) 100.0 -3.10 [ -5.26, -0.94 ]

Subtotal (95% CI) 100.0 -3.10 [ -5.26, -0.94 ]




Leibing 2002 -6.80 (2.96) 100.0 -6.80 [ -12.60, -1.00 ]

Subtotal (95% CI) 100.0 -6.80 [ -12.60, -1.00 ]

-100 -50 0 50 100

Favours acup + inter Favours interv alone (Continued . . . )



(. . . Continued)


95% CI (%) 95% CI







-100 -50 0 50 100



> 3 months), Outcome 06 global measure




Study Acup + interv Intervention alone Relative Risk (Random) Relative Risk (Random)

n/N n/N 95% CI 95% CI


Molsberger 2002 52/62 31/55 1.49 [ 1.15, 1.92 ]


Molsberger 2002 36/49 11/37 2.47 [ 1.46, 4.17 ]



0.1 0.2 0.5 1 2 5 10

favours interv alone favours acup + inter




> 3 months), Outcome 07 back specific functional status (lower scores mean better). Ex: RDQ, Oswestry and

Aberdeen



Outcome: 07 back specific functional status (lower scores mean better). Ex: RDQ, Oswestry and Aberdeen




Leibing 2002 35 11.30 (15.00) 39 22.30 (7.80) 43.0 -0.93 [ -1.41, -0.44 ]

Meng 2003 24 6.50 (4.00) 23 11.20 (4.80) 26.5 -1.05 [ -1.66, -0.43 ]

Yeung 2003 26 20.02 (10.47) 26 30.82 (13.03) 30.4 -0.90 [ -1.47, -0.33 ]

Subtotal (95% CI) 85 88 100.0 -0.95 [ -1.27, -0.63 ]




Meng 2003 24 6.30 (4.40) 23 11.40 (4.80) 46.0 -1.09 [ -1.71, -0.47 ]

Yeung 2003 26 20.36 (13.06) 26 32.48 (15.31) 54.0 -0.84 [ -1.41, -0.27 ]

Subtotal (95% CI) 50 49 100.0 -0.95 [ -1.37, -0.54 ]




Leibing 2002 33 16.20 (12.50) 30 22.60 (10.00) 54.7 -0.56 [ -1.06, -0.05 ]

Yeung 2003 26 19.36 (10.12) 26 25.82 (13.11) 45.3 -0.54 [ -1.10, 0.01 ]

Subtotal (95% CI) 59 56 100.0 -0.55 [ -0.92, -0.18 ]







-4 -2 0 2 4

favours acup + inter favours interv alone




> 3 months), Outcome 08 spine range of motion: difference between within group changes



Outcome: 08 spine range of motion: difference between within group changes

Study difference between (SE) difference between (Random) difference between (Random)

95% CI 95% CI


Leibing 2002 -4.50 (3.26) -4.50 [ -10.89, 1.89 ]



Leibing 2002 -4.70 (3.19) -4.70 [ -10.95, 1.55 ]


-10 -5 0 5 10



> 3 months), Outcome 09 Side effects / Complications




Study Acupuncture + interv Intervention alone Risk Difference (Random) Risk Difference (Random)

n/N n/N 95% CI 95% CI


Leibing 2002 3/35 0/39 0.09 [ -0.02, 0.19 ]

Yeung 2003 0/26 0/26 0.0 [ -0.07, 0.07 ]




-1 -0.5 0 0.5 1




Documents

Furlan AD, van Tulder MW, Cherkin DC, Tsukayama H, Lao L ...thedryneedlinginstitute.net/wp-content/uploads/2012/02/Dry... · Analysis 01.03. Comparison 01 acupuncture versusno treatment